JP7355316B2 - Mobile object management system, mobile object management device, and mobile object management method - Google Patents

Description

Article 30, Paragraph 2 of the Patent Act applies Presented at "The IEEE/RSJ International Conference on Intelligent Robots and Systems" held in Madrid, Spain for 5 days from October 1, 2018 [Published] Things] 2018 Presented at the Tokyo Ginza Rotary Club (Chuo-ku, Tokyo) on December 19th at the "Civil Engineering Revolution by Utilizing Cutting-edge Technology: Collaborative Development Across the Sea between Top American Universities and Niigata Companies"

The present invention relates to a technique for managing a movement plan of a moving object.

Unmanned aerial vehicles (UAVs), which are a type of mobile object, have improved their operability due to improvements in technology related to attitude control, flight control, etc. In addition, the number of small and low-priced unmanned aerial vehicles is increasing. Against this background, in recent years, attempts have been made to utilize it in various industrial fields.
For example, in the field of architecture and civil engineering, unmanned aerial vehicles are sometimes used to inspect structures such as bridges and buildings (see Patent Document 1).

JP2019-127245A

When unmanned aerial vehicles are used to inspect large structures, the scope of inspection is wide-ranging. Therefore, the flight distance required for the unmanned aircraft that conducts the inspection also increases. Therefore, it is conceivable that an unmanned aircraft may malfunction due to some kind of abnormality occurring during inspection due to, for example, a load on the aircraft body due to a long flight distance. Therefore, at inspection work sites using unmanned aerial vehicles, it is desired to be able to continue performing efficient and appropriate inspections even if the work is interrupted (for some reason) due to a malfunction of the unmanned aerial vehicle or the like.

An aspect of the present invention aims to realize efficient and appropriate continuous inspection even if the work is interrupted in the inspection work of a structure using two different moving bodies.

A management system for a moving body, which is one aspect of the present invention, is a system for carrying out inspection work on a structure whose inspection range is wide, using a first moving body and a second moving body. The first mobile body is equipped with a battery as a power source for supplying power to the drive device, and performs inspection work. The second moving body receives and records the result data of the inspection work from the first moving body.
This management system includes a planning section, a determination section, and a replanning section. In order for the first moving body to perform all inspection work on the structure, the planning department calculates the shortest travel route of the first moving body for all inspection sections, and A travel route for the first moving object is planned. The determination unit determines whether or not to interrupt the inspection work. When it is determined that the inspection work is to be suspended, the replanning unit calculates the shortest movement of the first moving body for the remaining uninspected sections excluding the inspected sections that were carried out before the inspection work was suspended in all inspection sections. The route is recalculated, and the travel route of the first mobile body after restarting the inspection work is replanned.

According to one aspect of the present invention, even when work using two different moving bodies is interrupted, efficient and appropriate inspections can be continued.

FIG. 1 is a diagram showing an example of a management system according to the first embodiment. FIG. 2 is a diagram showing an example of a computer system. FIG. 3 is a schematic diagram showing the flow of waterway inspection work according to the first embodiment. FIG. 4 is a diagram showing processing of the management system according to the first embodiment. FIG. 5 is a diagram showing how the unmanned aircraft according to the first embodiment takes off. FIG. 6 is a diagram showing how a waterway is inspected by an unmanned aircraft according to the first embodiment. FIG. 7 is a diagram illustrating an example of a flight path during waterway inspection by the unmanned aircraft according to the first embodiment. FIG. 8 is a diagram showing how the unmanned aircraft according to the first embodiment lands. FIG. 9 is a diagram showing an overview of the movement planning function (advance planning function) according to the first embodiment. FIG. 10 is a diagram showing an overview of the movement planning function (replanning function) according to the first embodiment. FIG. 11 is a functional block diagram of the management system (management device and unmanned aircraft) according to the first embodiment. FIG. 12A is a diagram illustrating an example of graph generation according to the first embodiment. FIG. 12B is a diagram illustrating an example of graph generation according to the first embodiment. FIG. 12C is a diagram illustrating an example of graph generation according to the first embodiment. FIG. 13 is a diagram illustrating an example of subgraph generation according to the first embodiment. FIG. 14 is a flowchart showing calculation processing for generating a subgraph according to the first embodiment. FIG. 15 is a diagram illustrating an example of the shortest path problem for vehicles and unmanned aerial vehicles for the subgraph according to the first embodiment. FIG. 16 is a flowchart showing calculation processing for solving the shortest route problem for vehicles and unmanned aircraft according to the first embodiment. FIG. 17 is a diagram showing an example in which the shortest route problem for vehicles between subgraphs according to the first embodiment is replaced with an asymmetric traveling salesman problem. FIG. 18 is a flowchart of calculation processing for solving the shortest route problem for vehicles between subgraphs according to the first embodiment. FIG. 19 is a flowchart showing the mobile object management process (pre-planning process) according to the first embodiment. FIG. 20 is a flowchart showing the mobile object management process (replanning process) according to the first embodiment. FIG. 21 is a functional block diagram of a management device according to a first modification. FIG. 22 is a functional block diagram of a management device according to a second modification.

First, in the practical embodiment of the present disclosure, an example will be described in which a mobile body management system is utilized for "waterway inspection work." In Japan, irrigation canals that are tens of kilometers long are inspected manually. Such manual inspection methods are insufficient to cope with the increasing demand for periodic inspections due to aging and the limited seasonal periods suitable for inspections. Therefore, for the purpose of shortening inspection time (improving work efficiency), automatic inspection using a combination of different moving objects, such as an unmanned aircraft and a ground vehicle, is effective as an alternative to manual inspection. From the viewpoint of such industrial application effects, the embodiment of the present disclosure will cite an example of application to waterway inspection work.

Hereinafter, embodiments according to the present invention will be described with reference to the drawings. Note that the present invention is not limited to the contents of the embodiments. The configuration of the present invention can be modified as appropriate without departing from the technical idea of the present disclosure.

<First embodiment>
[System configuration]
FIG. 1 is a diagram showing an example of a mobile body management system 1 according to the present embodiment. As shown in FIG. 1, at least two different moving bodies are used at a work site. Specifically, they are the vehicle 2 (second moving object) and the unmanned aerial vehicle 3 (first moving object).

The vehicle 2 is a moving body that travels under the driving operation of a driver Dr. The vehicle 2 is a base vehicle that moves on roads around the waterway and functions as a base for inspection work so that the manager Op on board can monitor the work performed by the unmanned aerial vehicle 3. Note that the vehicle 2 may travel by remote control or autonomously according to a movement plan (driving plan) to be described later. In the following description, the vehicle 2 will be referred to as the base vehicle 2 for convenience. The unmanned aircraft 3 is a moving object that flies unmanned. As described above, the unmanned aircraft 3 enters the waterway to be inspected, flies within the waterway for a predetermined distance or for a predetermined time, and inspects the deterioration state of the waterway wall and the like. The unmanned aircraft 3 flies autonomously according to a movement plan (flight plan) that will be described later. In the following description, the unmanned aerial vehicle 3 will be referred to as a drone 3 for convenience.

The management system 1 includes a management device 4, an input device 5, an output device 6, a communication device 7, and a position sensor 8. The management device 4, the input device 5, the output device 6, the communication device 7, and the position sensor 8 are installed in the base vehicle 2, for example. The communication device 7 performs communication between the management device 4 and the drone 3. The management device 4 includes a wireless communication device 9, and the drone 3 includes a wireless communication device 34. Therefore, the management device 4 and the drone 3 communicate wirelessly via the communication device 7. The communication device 7 communicates between the management device 4 and the drone 3 using, for example, an IP address routing function.

The input device 5 receives input operations from the administrator Op, for example, and generates input data. Input data generated by the input device 5 is output to the management device 4. The input device 5 is at least one of a computer keyboard, a button, a switch, a touch panel, and the like.

The output device 6 receives an output signal from the management device 4 and outputs data, thereby providing information to, for example, the administrator Op. The output device 6 is at least one of a display device capable of displaying display data, an audio output device capable of outputting audio, a printing device capable of outputting printed matter, and the like. Note that the display device includes a flat panel display such as a liquid crystal display (LCD) or an organic electroluminescence display (OELD).

The position sensor 8 detects the position of the base vehicle 2. The position sensor 8 detects the position of the base vehicle 2 using the Global Navigation Satellite System (GNSS). The global navigation satellite system includes a global positioning system (GPS) and the like. The global navigation satellite system detects the absolute position of the base vehicle 2 defined by, for example, coordinate data of latitude, longitude, and altitude. A global navigation satellite system detects the position of the base vehicle 2 defined in the global coordinate system. A global coordinate system is a coordinate system fixed to the earth. The position sensor 8 includes a GPS receiver and the like, and detects the absolute position (coordinates) of the base vehicle 2.

The drone 3 includes a flight control device 30, a flight device 31, a main body 32, a position sensor 33, a wireless communication device 34, a power source 35, and a measuring device 36.

The drone 3 flies by rotating the propeller 31P. Therefore, the flight device 31 includes a propeller 31P, a drive device 31D, and the like. The drive device 31D generates a driving force for rotating the propeller 31P. The drive device 31D includes an electric motor and the like. The flight control device 30 is called a flight controller, and performs predetermined calculations based on measurement data from a measurement device 36, which will be described later, to control the attitude and flight of the aircraft. The flight control device 30 controls the aircraft by transmitting a control signal to the drive device 31D based on the calculation result. The main body 32 is supported by the flight device 31. The position sensor 33 detects the position of the drone 3. The position sensor 33 includes a GPS receiver and the like, and detects the absolute position (coordinates) of the drone 3. The wireless communication device 34 is capable of wireless communication with the communication device 7 mounted on the base vehicle 2.

The drone 3 has a power source 35 that supplies power to an electric motor. The power source 35 includes a rechargeable battery (battery) and the like. In the following description, the power source 35 will be referred to as a battery 35 for convenience.

The measurement device 36 includes an imaging device 37, an inertial measurement unit (IMU), and the like, and senses the surroundings of the drone 3 and the drone itself, and measures the flight environment and flight state. The imaging device 37 images a subject and acquires image data. The imaging device 37 includes an optical device and an image sensor. Optical devices include optical lenses and the like. The image sensor includes a CCD (Couple Charged Device) image sensor, a CMOS (Complementary Metal Oxide Semiconductor) image sensor, or the like. The inertial measurement device (not shown) includes an angular velocity meter (gyro) that detects the angular velocity (rotational motion) of the moving body, an accelerometer that detects the acceleration (linear motion), etc., and measures the movement of the moving body. In addition to the accelerometer and the angular velocity meter, the inertial measurement device may include a magnetometer (three axes), a temperature sensor, a processor, and the like.

[Basic configuration of computer system]
FIG. 2 is a diagram showing an example of the computer system 100 according to this embodiment.
Each of the management device 4 and flight control device 30 described above includes a computer system 100 and the like. Computer system 100 includes a processor 101, memory 102, storage 103, and interface 104. The processor 101 is, for example, a CPU (Central Processing Unit). The memory 102 includes nonvolatile memory such as ROM (Read Only Memory), volatile memory such as RAM (Random Access Memory), and the like. The storage 103 includes an HDD (Hard Disk Drive), an SSD (Solid State Drive), and the like. The interface 104 includes input/output circuits and the like. The functions of the management device 4 and the flight control device 30 described above are stored in the storage 103 as a program. The processor 101 reads the program from the storage 103, expands it into the memory 102, and executes processing according to the program. Note that the program may be distributed to the computer system 100 via a public line.

[Waterway inspection work]
FIG. 3 is a schematic diagram showing the flow of waterway inspection work according to the present embodiment. As shown in Figure 3, the waterway inspection work is divided into several work steps. First, several persons in charge, including at least the site supervisor who is the manager, go to the site where the inspection work is to be carried out and conduct a preliminary investigation (step S1). The person in charge considers the overall work plan regarding the inspection based on the results of the preliminary investigation (step S2). The work plan considered at this time includes a movement plan for the base vehicle 2 and the drone 3, which will be described later. Further, the work plan is performed using the functions of the management device 4.

The person in charge returns to the site using the base vehicle 2 carrying the management device 4 and the drone 3 and performs inspection work (step S3). At this time, the drone 3 starts flying toward the waterway to be inspected and inspects the waterway according to a predetermined work plan. The base vehicle 2 starts traveling on the roads around the waterway according to the work plan. As a result, the manager Op on board monitors the drone 3 being inspected and manages the implementation work.

After the inspection, the person in charge processes the measurement data of the waterway (step S4). In the data processing referred to here, the reliability of measurement data is confirmed, data is analyzed using a predetermined tool, and a report is created based on the analysis results.

FIG. 4 is a diagram showing the processing of the management system 1 according to this embodiment. The process shown in FIG. 4 is the details of the process executed in step S3 described above. As shown in FIG. 4, in the management system 1, the following process is executed between the base vehicle 2 and the drone 3 during field work.

In the base vehicle 2, the manager Op on board instructs the drone 3 to take off via the management device 4 (step S21). After that, the base vehicle 2 starts traveling, and the administrator Op monitors the drone 3 (step S22). The base vehicle 2 travels on roads around the waterway as the driver Dr performs driving operations according to a travel plan planned in advance (step S23).

When the base vehicle 2 arrives at the body recovery position of the drone 3 according to the movement plan (step S24), the administrator Op recovers the body of the drone 3 (step S25). Then, during the inspection, the administrator Op replaces the battery 35 of the recovered drone 3 (step S26), and installs the drone 3 after replacing the battery 35 at a predetermined takeoff position. At this time, if the takeoff position is far from the current aircraft recovery position by a predetermined distance or more, the base vehicle 2 carries the drone 3 after replacing the battery 35 to the takeoff position.

When the drone 3 receives the takeoff instruction, the flight control device 30 controls the drive device 31D, rotates the propeller 31P, and takes off (step S31). Then, as shown in FIG. 5, the drone 3 enters the waterway C from the takeoff position and moves to the inspection start position SP (step S32).

For example, as shown in FIG. 6, the drone 3 autonomously flies and inspects the waterway C according to a movement plan planned in advance (step S33). At this time, the drone 3 has an object detection region DR, and flies so as not to come into contact with objects around the aircraft body (for example, the right wall RW and the left wall LW with respect to the direction of travel). In addition, the drone 3 uses the imaging device 37 to image both the left and right walls RW, LW in the waterway C within the same angle of view IA (hereinafter referred to as "imaging range"), and transfers the captured images to the base vehicle 2 (management device 4). Thereby, the administrator Op determines the deterioration state of the waterway C by using the management device 4 to analyze the time series data of the accumulated captured images. Note that if the imaging range IA by the imaging device 37 does not include the right and left walls RW and LW of the waterway C, the drone 3 flies in a zigzag pattern within the waterway C, as shown in FIG. 7, for example. Good too. Thereby, the drone 3 can image the entire range of the walls RW and LW to be inspected, even if, for example, the width of the waterway C is much wider than the width of the imaging range IA.

When the drone 3 arrives at the inspection end position E according to the movement plan (step S34), as shown in FIG. 8, the drone 3 raises the flight position of the aircraft from the inspection end position EP to a predetermined height. As a result, after leaving the waterway C, the drone 3 moves to the body recovery position and lands (step S35). As described above, after that, the landed drone 3 is recovered by the administrator Op, and the battery 35 is replaced.

[Functional configuration]
In large-scale agricultural areas, the work area of waterway C, which is the subject of inspection, is wide. Specifically, in large-scale agricultural areas, several tens of kilometers of waterways C extend across several ^tens of kilometers of farmland. On the other hand, the drone 3 normally used in the industrial field can fly for about 20 to 40 minutes in a flight area within about 7 km within which communication is possible from the base where the administrator Op is located.

In this way, the waterway C to be inspected exceeds the range in which the drone 3 can continue to fly based on the capacity of the onboard battery 35. Therefore, during the entire inspection work, it is necessary to repeatedly perform the inspection work while replacing the battery 35 of the drone 3 multiple times. Furthermore, the length of the waterway C places a greater than expected load on the drone 3 that performs the inspection. Therefore, even if the work is interrupted, it is necessary to restart the inspection work using the drone 3 and carry out the remaining work efficiently and appropriately.

Therefore, in the management system 1 according to the present embodiment, the management device 4 performs the following processing so that the battery 35 can be replaced at an appropriate timing and repeated inspection work can be carried out efficiently. Specifically, the management device 4 optimizes the movement route of the entire inspection work in the base vehicle 2 and the drone 3 (plans the movement route). At this time, the management device 4 calculates and determines the optimal travel route based on the continuous flight range according to the capacity of the battery 35, etc. Then, the management device 4 reflects the optimized movement route data in the movement plan data of the base vehicle 2 and the drone 3.

Furthermore, in the management system 1 according to the present embodiment, the management device 4 performs the following processing so that even if the work is interrupted, efficient and appropriate inspections can be continued. Specifically, the management device 4 re-optimizes the movement route of the entire inspection work in the base vehicle 2 and the drone 3 (re-plans the movement route) based on the position where the work is interrupted. At this time, the management device 4 determines the optimal movement routes for each of the base vehicle 2 and the drone 3 for the remaining inspection areas (uninspected areas) excluding the inspected areas where work has been carried out according to the movement route before the suspension of work. Recalculate and decide. Then, the management device 4 reflects the optimized movement route data again in the movement plan data of the base vehicle 2 and the drone 3.

The movement planning function of the management system 1 according to this embodiment will be described below. As described above, the movement planning function according to the present embodiment is mainly divided into two functions: a planning function before work execution and a replanning function during work.
[Advance planning function]
As shown in FIG. 9(A), the movement planning function according to the present embodiment first creates a waterway path P _C representing the entire waterway C to be inspected and a road path _PR representing roads around the waterway C on a map. Extracted from data 441. Next, as shown in FIG. 9(B), the movement planning function divides the waterway path _PC included in the extracted data according to the range in which the drone 3 can continue to fly without replacing the battery 35, and divides it into multiple divisions. A waterway path DP _Cn (inspection section corresponding to the distance or time that can be continuously flown among all inspection targets) is generated. Next, as shown in FIG. 9C, the movement planning function optimizes each movement route of the base vehicle 2 and the drone 3 for each divided waterway path DP _C (divided inspection section). In other words, the movement planning function determines the first movement route (the movement distance or movement time of the base vehicle 2 and the drone 3) in the divided inspection section, in which the base vehicle 2 and the drone 3 can maintain a communicable positional relationship. A first movement route determination process to be optimized is executed. Next, the movement planning function executes optimization of the movement route of the base vehicle 2 that connects each divided waterway path DP _Cn , as shown in FIG. 9(D). In other words, the movement planning function executes the second movement route determination process that optimizes the second movement route (the movement distance or movement time when the base vehicle 2 carries the drone 3) that connects each divided inspection section. do.

In this way, the movement planning function according to the present embodiment allows the drone 3 to move along the waterway path PC representing all the waterways _C to be inspected within the range in which the drone 3 can continue to fly without replacing the battery 35 (distance or time over which the drone 3 can continue to fly). : 20 to 40 minutes). Then, the movement planning function optimizes the movement route (first movement route) within the path for the plurality of divided waterway paths DP _Cn . Then, the movement planning function optimizes a movement route (second movement route) that connects each divided waterway path DP _Cn whose movement route has been optimized for each path. Thereby, the management system 1 (management device 4) according to the present embodiment can appropriately determine the travel route according to the capacity of the battery 35. That is, the management system 1 can create a movement plan for the base vehicle 2 and the drone 3 (two different types of moving objects) that replaces the battery 35 at an appropriate timing and implements efficient work.

[Replanning function]
FIG. 10 is a diagram showing an overview of the movement planning function (replanning function) according to this embodiment. As shown in FIG. 10, the movement planning function according to _the present embodiment determines the position IP(t) (hereinafter referred to as Check the time t when the work was interrupted (hereinafter referred to as the "work interruption time"). Next, based on the work interruption position IP and/or the work interruption time _t , the movement planning function selects an inspected section _ST1 (Fig. (dotted line in the middle). Then, the movement planning function deletes the path corresponding to the identified inspected section _ST1 from the divided waterway path _DPC . The movement planning function recalculates and determines the optimal movement route for the base vehicle 2 and the drone 3 for the uninspected section ST ₂ (solid line in the figure) in the divided waterway path DP _C. In other words, the movement planning function excludes the inspected section _ST1 before the work interruption, which is specified by the work interruption time t, from the movement route recalculation target.

As a result, the management system 1 (management device 4) according to the present embodiment quickly determines the optimal movement route for the base vehicle 2 and the drone 3 for the remaining divided waterway paths DP _C including the uninspected section ST ₂ after the work is interrupted. can be determined. That is, the management system 1 can create a movement plan for the base vehicle 2 and the drone 3 (two different types of moving objects) that realizes efficient and appropriate continuation of inspection work even if the work is interrupted.

FIG. 11 is a functional block diagram of the management system 1 according to this embodiment. The management device 4 and drone 3 included in the management system 1 according to this embodiment have functions as shown in FIG. 11.

[Management device]
The management device 4 includes a transmitting/receiving section 40, a graph processing section 41, a route planning section 42, a mobile object control section 43, a storage section 44, a work interruption determination section 45, a replanning section 46, etc., and each functional section cooperates. It provides a predetermined function by operating as follows. The transmitting/receiving unit 40 transmits and receives data to and from other devices other than the management device 4, such as the drone 3. The storage unit 44 is a predetermined storage area that stores various data necessary for the management device 4 to execute the movement planning function. Note that the storage unit 44 is a function that the memory 102 and/or the storage 103 shown in FIG. 2 have.

Various data according to this embodiment include map data 441, parameter data 442, movement plan data 443, measurement data 444, and the like. The map data 441 is map data including the entire waterway C to be inspected and roads around it. The map data 441 is obtained in advance, for example, at the stage of a field survey. The parameter data 442 is data indicating various parameter values for determining the movement route of the base vehicle 2 and/or the drone 3 in the movement plan. Parameter data 442 mainly includes information regarding each of base vehicle 2 and drone 3. Parameter values regarding the base vehicle 2 include, for example, the total number of vehicles used for inspection work, the communication distance by the communication device 7 (the distance at which wireless communication is possible from the vehicle), and the like. Parameter values related to the drone 3 include, for example, the total number of drones used for inspection work, the continuous flight time without replacing the battery 35, and the flight speed. Additionally, the parameter data 442 may further include information regarding external factors that affect the inspection work. The parameter values of the external factors include, for example, traffic information on the road on which the base vehicle 2 is scheduled to travel, weather information on the area where the inspection work is to be carried out, information on setting a no-fly zone based on the Civil Aeronautics Act, and the like. Note that, as the parameter value of the external factor, the latest information obtained via a public line (Internet), for example, may be stored. On the other hand, parameter values other than external factors may be stored in advance, for example, determined based on the results of a field survey. The movement plan data 443 is data indicating a movement plan of moving bodies for carrying out inspection work on the waterway C using two different types of moving bodies. The movement plan data 443 includes optimal movement route data (movement route optimization data) for the base vehicle 2 and/or the drone 3, and the like. The movement plan data 443 is generated by the route planning section 42, which will be described later, and is stored in the storage section 44. The measurement data 444 is data indicating the inspection results of the waterway C, and is measurement data of various sensors. The measurement data 444 includes imaging data of the imaging device 37 and the like. The measurement data 444 is transmitted from the drone 3 to the management device 4 via a predetermined communication line. As a result, measurement data 444 is received via the transmitting/receiving section 40 and stored in the storage section 44. At this time, the measurement data 444 is periodically transmitted from the drone 3 and stored in the storage unit 44 as time-series data.

[Graph processing function]
The graph processing unit 41 generates data that can be used in the movement planning function from map data 441 that includes the entire waterway C to be inspected and roads around it, and performs predetermined data processing. In this embodiment, graph data, which is one type of mathematical model, is generated. Note that graph data here is a data structure that is expressed by a set of nodes (nodes/vertices) and a set of edges (branches/edges), and represents a path between nodes based on the connection relationship of each edge. be.

The graph processing section 41 includes a graph generation section 411 and an edge division section 412.

FIG. 12 is a diagram showing an example of graph generation according to this embodiment. The graph generation unit 411 generates graph data GD from the map data 441. As shown in FIG. 12A, the graph generation unit 411 extracts a waterway path PC representing the entire waterway _C to be inspected from the map data 441 through predetermined image processing (for example, filter processing based on a predetermined feature amount). Furthermore, the graph generation unit 411 extracts a road path _PR representing roads around the waterway C from the map data 441 by performing predetermined image processing. The graph generation unit 411 generates graph data GD of the waterway C from the extracted data (path data) of the waterway C, as shown in FIG. 12B. For example, the graph generation unit 411 inserts nodes into the extracted waterway paths _PC at connection points between the paths. As a result, the graph generation unit 411 generates path data between nodes (between nodes ND _n and ND _n+1 ) by a plurality of nodes ND _n (1≦n≦a) and a plurality of edges EG _n (1≦n≦a). Graph data GD having a data structure representing a route between (intervals) is generated. That is, the graph generation unit 411 converts the map data 441 into graph data GD via the path data.

As shown in FIG. 12C, the edge dividing unit 412 divides each edge EG _n constituting the graph data GD into equal intervals to reconstruct the graph data GD. As mentioned above, the working range of waterway C is wide. Therefore, the work target range corresponding to one edge ED may be wider than the range in which the drone 3 can perform inspection work without replacing the battery 35. Specifically, the distance or time of the work target corresponding to one edge ED may be longer than the distance or time at which the drone 3 can perform the inspection work without replacing the battery 35. Therefore, in this embodiment, the edge dividing unit 412 divides one edge EG at equal intervals. A path planning unit 42, which will be described later, connects and combines the plurality of divided edges SEG _nm to generate subgraph data corresponding to the inspection section. Thereby, in this embodiment, it is possible to flexibly configure an inspection section in which the drone 3 can perform inspection work without replacing the battery 35.

Note that the division intervals d1 and d2 of the edge EG (distance or time corresponding to the division edge SEG) need to be shorter than the distance or time that the drone 3 can continue to fly without replacing the battery 35. On the other hand, if the division intervals d1 and d2 are too small (if the division granularity is too fine), the amount of calculations related to determining the movement route, which will be described later, will increase (the calculation cost will increase). Therefore, the division intervals d1 and d2 of the edge EG are determined by performing a simulation in advance while considering the work plan, and are optimized intervals (distance or time in advance) based on the distance or time that can be continuously flown and the amount of calculation. It is desirable to use the adjusted value).

[Movement planning function (pre-planning function)]
The route planning unit 42 optimizes the travel route for the drone 3 to fly and inspect on the waterway path _PC representing the waterway C. The route planning unit 42 optimizes the travel route along which the base vehicle 2 travels on the road path _PR of the roads around the waterway C and follows the flight of the drone 3. At this time, the route planning unit 42 calculates and determines the optimal travel route based on the possible continuous flight time and flight speed according to the capacity of the battery 35. In this way, the movement planning unit 42 optimizes the movement route of the entire inspection work in the base vehicle 2 and the drone 3, and reflects the data of the optimized movement route in the movement plan data 443 of the base vehicle 2 and the drone 3. . As a result, in the inspection work of the waterway C, the battery 35 of the drone 3 can be replaced at an appropriate timing, and the optimization of two different types of mobile objects can be realized to realize efficient work using the base vehicle 2 and the drone 3. A travel plan will be created.

The route planning unit 42 includes a subgraph generation unit 421, a first movement route determination unit 422, and a second movement route determination unit 423.

FIG. 13 is a diagram illustrating an example of subgraph generation according to this embodiment. The subgraph generation unit 421 generates subgraph data SGD representing an inspection section in which the drone 3 can perform inspection work without replacing the battery 35. As shown in FIG. 13, the subgraph generation unit 421 generates a plurality of subgraph data SGD _n (1≦n≦c) from the graph data RGD that is obtained by dividing each edge EG _n at equal intervals and reconstructing the graph data RGD.

In order to generate subgraph data SGD, it is necessary to determine the number of generated subgraph data SGD (total number of subgraphs) and the number of edges forming subgraph data SGD (total number of edges of one subgraph). Specifically, in generating the subgraph data SGD, the battery 35 of the drone 3 is replaced and the time required to travel between each divided waterway path DP _Cn of the waterway C is minimized in the movement route of the entire inspection work of the waterway C. It is necessary to determine the number of generated subgraph data SGD and the number of edges such that . Therefore, in this embodiment, this problem (minimization problem of the number of generated subgraphs) is mathematically modeled (the problem is formulated), and a method of solving the problem by mathematical optimization (a problem solving method) is applied. Thereby, the subgraph generation unit 421 calculates the number of generated subgraph data SGD and the number of edges, and generates a plurality of subgraph data SGD _n based on the calculation results.

A mathematical model is composed of variables, constraints, and objective functions. In this embodiment, the above problem is a quadratic programming problem (QP) in which the objective function is a quadratic expression and the constraint is a linear expression. Furthermore, in this embodiment, among the quadratic programming problems, a mixed integer quadratic programming problem (MIQP: Mixed Integer Quadratic Programming) is used when integers are included in variables.

FIG. 14 is a flowchart showing calculation processing for generating a subgraph according to this embodiment. The subgraph generation unit 421 mathematically models the above problem as a mixed integer quadratic programming problem, and executes calculation processing as shown in FIG. Thereby, the subgraph generation unit 421 generates a plurality of subgraph data SGD _n based on the calculation results of the number of generated subgraph data SGD and the number of edges.

（変数リスト）
Ｋ：ドローンの総数；
Ｍ：完全に充電されたバッテリによってドローンが点検作業を実施可能な範囲に相当するエッジｅの数（ドローンがバッテリの交換なしで継続飛行可能な範囲に相当するエッジｅの数）；
Ｎｅ：再構成後のグラフのエッジｅの総数；及び
ＫＭ：サブグラフを構成するエッジｅの最大数である。 As shown in FIG. 14, the subgraph generation unit 421 sets an initial value _S0 of the number S of generated subgraph data SGD (step S41). The initial value S ₀ is expressed by the following formula [1].

(variable list)
K: Total number of drones;
M: the number of edges e corresponding to the range in which the drone can perform inspection work with a fully charged battery (the number of edges e corresponding to the range in which the drone can continue to fly without replacing the battery);
Ne: total number of edges e in the graph after reconstruction; and KM: maximum number of edges e constituting the subgraph.

Note that the variable values K and M are each parameter value included in the parameter data 442 stored in the storage unit 44 or a value calculated from the parameter value. For example, the variable value K is the value of the total number of drones 3 used for inspection work. Further, the variable value M is a value calculated for the edge EG in the graph data RGD after reconfiguration, based on the value of the flight time and the flight speed during which the drone 3 can continue to fly without replacing the battery 35. . In this way, the subgraph generation unit 421 determines the number S of generated subgraph data SGD and the number of edges, taking into consideration the total number of drones 3 used during inspection work and the possible continuous flight range according to the capacity of the battery 35. There is.

After setting the initial value _S0 , the subgraph generation unit 421 uses the following equations [2] to [6] (objective function and formulated constraints), which are mathematical models of the mixed integer quadratic programming problem. Then, the problem of minimizing the number of generated subgraphs is solved (step S42).
(objective function)

(formulated constraints)

(variable list)
x _si ∈{0,1}: decision variable indicating whether node i is included in subgraph s;
x _sj ∈{0,1}: a decision variable that indicates whether node j is included in subgraph s; and w _se ∈{0,1}: a decision variable that indicates whether edge e is included in subgraph s. It is.

(constraint list)
Constraint [3]: Each edge e must belong to only one subgraph s;
Constraint [4]: Let the total number of drones K be the lower limit of the number E of edges of all subgraphs s, and let the maximum number KM of subgraphs s be the upper limit of the number E of edges of all subgraphs s;
Constraint [5]: It is necessary to connect each subgraph s; and Constraint [6]: When edge e _ij is included in subgraph s, both node i and node j are included in subgraph s.

The subgraph generation unit 421 determines whether calculation of the mathematical model of the mixed integer quadratic programming problem is executable based on the above equations [2] to [6] (step S43). In other words, the subgraph generating unit 421 determines whether the problem of minimizing the number of generated graphs S has been solved. If the subgraph generation unit 421 determines that it is not executable (step S43: NO), it adds 1 to the generation number S (step S44). On the other hand, if the subgraph generation unit 421 determines that it is executable (step S43: YES), it ends the process. In this way, the subgraph generation unit 421 performs the iterative process until the mixed integer quadratic programming problem is solved. Then, the subgraph generation unit 421 increases the number S of generated subgraph data SGD by +1 from the initial value _S0 . As a result, the subgraph generation unit 421 generates a plurality of subgraph data based on the number of generated S (total number of subgraphs) and the number of edges E (total number of edges of one subgraph) at the time when the mixed integer quadratic programming problem is solved. Generate SGD _n .

After the subgraph data SGD is generated, it is necessary to determine an optimal movement route for the drone 3 to perform inspection work within the divided waterway path _DPC of the waterway C corresponding to the data. Therefore, the first movement route determining unit 422 optimizes each movement route (first movement route) of the base vehicle 2 and the drone 3 with respect to one generated subgraph data SGD. Specifically, the first movement route determination unit 422 determines the movement path of the drone 3 that passes through a plurality of nodes ND _n and a plurality of edges EG _n in one subgraph data SGD in the shortest possible time. Furthermore, the first movement route determining unit 422 determines, with respect to the movement path of the drone 3, the movement path of the base vehicle 2 located at a distance that allows communication with the moving drone 3 via the communication device 7. Therefore, in this embodiment, as in the case of subgraph generation, this problem is mathematically modeled and a method of solving the problem by mathematical optimization is applied.

FIG. 15 is a diagram illustrating an example of the shortest path problem for the base vehicle 2 and the drone 3 for the subgraph according to the present embodiment. The example shown in FIG. 15 assumes a case where one drone 3 inspects a divided waterway path DP _C of a waterway C expressed by one subgraph data SGD. Further, around the divided waterway path _DPC , there is a road (road path _PR ) on which the base vehicle 3 can travel. A communicable area CR with the base vehicle 2 (communication device 7 mounted on the vehicle) exists around the drone 3. In this way, when the total number of base vehicles 2 and the total number of drones 3 used in inspection work are determined for one subgraph data SGD, the shortest path problem for base vehicles 2 and drones 3 (in the subgraph can be mathematically modeled as a mixed integer quadratic programming problem.

FIG. 16 is a flowchart showing calculation processing for solving the shortest route problem for the base vehicle 2 and the drone 3 according to this embodiment. The first movement route determining unit 422 mathematically models the above problem as a mixed integer quadratic programming problem, and executes calculation processing as shown in FIG. 16. Thereby, the first movement route determination unit 422 determines the shortest movement route of the drone 3 for one subgraph data SGD, and determines the shortest movement path of the base vehicle 2 according to the shortest movement path of the drone 3. That is, the first movement route determining unit 422 determines a first movement route including the shortest movement routes of the base vehicle 2 and the drone 3 for one subgraph data SGD.

（変数リスト）
Ｎｓ：サブグラフｓにおけるエッジｅの総数である。 As shown in FIG. 16, the first movement route determination unit 422 sets an initial value T ₀ of the work planning target time (discrete time step) T for the divided waterway path DP _C of the waterway C expressed by the subgraph data SGD. (Step S51). The initial value T ₀ is expressed by the following formula [7].

(variable list)
Ns: Total number of edges e in subgraph s.

After setting the initial value T ₀ , the first movement route determination unit 422 calculates the following equations [8] to [14] (objective function and formulated constraints) by mathematically modeling the mixed integer quadratic programming problem. is used to solve the problem of minimizing the movement distance within the subgraph (step S52).
(objective function)

(formulated constraints)

(variable list)
K _car : total number of base vehicles;
x _kti ∈{0,1}: decision variable representing whether drone k is located at node i of the subgraph at time t;
y _kcarti ∈{0,1}: decision variable representing whether base vehicle kcar is located at node i of the subgraph at time t; and ω _ktkcar ∈{0,1}: drone k is located at base vehicle kcar at time t. a decision variable representing whether or not the location is within the communication coverage area;
w _ektd ε{0, 1}: a decision variable representing whether edge e is visited by drone k that starts an inspection flight from time t in direction d.
Note that d=1 for edge e _ij (node i < node j) means that when edge e _ij is inspected by drone k, the drone first reaches node i and then reaches node j. represents something. d=2 represents that the drone reaches node j first and then node i.
R: A transmission matrix composed of 0 and 1, and if the Euclidean distance between node i and node j is smaller than the maximum transmission distance of the drone, R (i, j) = 1;
Acanal: An adjacency matrix of a subgraph representing a divided channel path of a waterway composed of 0 and 1, and if node i and node j are adjacent, Acanal (i, j) = 1; and Aroad : This is an adjacency matrix of a subgraph representing a divided road path of a road consisting of 0 and 1, and when node i and node j are adjacent, Aroad (i, j)=1.

Note that the variable value ω _ktkcar is a value calculated from the parameter values included in the parameter data 442 stored in the storage unit 44. For example, the variable value ω _ktkcar is calculated by determining whether the drone 3 is located within the communication range CR of the base vehicle 2 based on the value of the distance at which the drone 3 can communicate with the base vehicle 2. This is the value. In this way, the first movement route determining unit 422 takes into account the communication area CR between the drone 3 and the base vehicle 2, and includes the shortest movement routes of the base vehicle 2 and the drone 3 for one subgraph data SGD. A first travel route is being determined.

(constraint list)
The constraint conditions mainly include constraints on the positional relationship between the base vehicle 2 and the drone 3 and constraints on the communicable range. Specifically, the details are as follows.
Constraint [9]: It is necessary to check every edge e once; and Constraint [10]: The decision variable w _ektd is a sufficient condition for the decision variable x _kti .
Specifically, w _ekt1 =1 is a sufficient condition for x _kti =1 and x _k(t+1)j =1, and w _ekt2 =1 is a sufficient condition for x _ktj =1 and x _k(t+1)i =1. This relationship is a sufficient condition.
Constraints [11], [12]: For all drones, at least one base vehicle kcar must always be located within the transmission range (within the communication range) of drone k;
Constraint [13]: All base vehicles Kcar and drone K must appear at only one node within a certain period of time; and Constraint [14]: Base vehicle kcar and drone K must appear at consecutive times. Move only to adjacent nodes in the process.

Note that the first movement route determining unit 422 takes into account the above constraints [9] to [14] in determining the first movement route including the shortest movement routes of the base vehicle 2 and the drone 3 for one subgraph data SGD. In addition, the following conditions are also assumed.
(Additional assumed conditions)
Assumption [1]: The end of the divided channel path of the channel represented by the subgraph needs to be inspected by a drone;
Assumption [2]: A certain amount of time is required for the drone to inspect the edges that make up the subgraph; and Assumption [3]: At the beginning and end of the inspection work, the drone will take a certain amount of time between the waterway and the base vehicle. The time required to fly the aircraft is considerably shorter than the inspection work time, so it does not need to be taken into consideration in the route determination process.

The reason for the above assumption [3] is as follows.
In order to acquire a clear captured image of the wall of the waterway C while moving during inspection work, the drone 3 needs to fly at a low speed (a speed that can maintain image quality). Therefore, in the route determination process, it is necessary to take this inspection work time into consideration. In comparison, when the drone 3 flies between the waterway C and the base vehicle 2, it may fly at a higher speed than during inspection work. Therefore, in the route determination process, there is no need to consider the flight time between the waterway C and the base vehicle 2.

The first movement route determination unit 422 determines whether calculation of the mathematical model of the mixed integer quadratic programming problem is executable based on the above equations [7] to [14] (step S53). In other words, the first movement route determining unit 422 determines whether the shortest route problem has been solved. If the first movement route determination unit 422 determines that it is not executable (step S53: NO), it adds 1 to the work planning target time (discrete time step) T (step S54). On the other hand, if the first moving route determining unit 422 determines that it is executable (step S53: YES), it ends the process. In this way, the first movement route determination unit 422 performs the iterative process until the mixed integer quadratic programming problem is solved. Then, the first movement route determination unit 422 increases the work planning target time T for the divided waterway path DP _C of the waterway C expressed by the subgraph data SGD from the initial value T ₀ by +1. The first travel route determining unit 422 determines the travel route at the time when the mixed integer quadratic programming problem is solved as the shortest travel route for the base vehicle 2 and the drone 3.

After determining the optimal travel route for inspecting the divided waterway paths DP _Cn of the waterway C, it is necessary to determine the shortest travel route for the base vehicle 2 to pass through all the divided waterway paths DP _Cn . Therefore, the second movement route determination unit 423 optimizes the movement route (second movement route) of the base vehicle 2 for all subgraph data SGD _n in which the movement route in the graph has been optimized. Specifically, the second travel route determination unit 423 determines the travel route of the base vehicle 2 that passes through a plurality of subgraphs in the shortest possible time.

FIG. 17 is a diagram illustrating an example in which the shortest path problem for base vehicles 2 between subgraphs according to the present embodiment is replaced with an asymmetric traveling salesman problem. In the example shown in FIG. 17, one subgraph SG _n expressing each divided waterway path DP _Cn of the waterway C is set as one node ND. As a result, the shortest path problem when passing through a plurality of subgraphs is replaced with an asymmetric traveling salesman problem (ATSP). More specifically, in this embodiment, new graph data NGD including a plurality of nodes ND corresponding to each subgraph SG _n is generated. In this embodiment, in order to allow the base vehicle 2 to freely move between nodes, the new graph NGD has a plurality of subgraphs SG _n connected to each other. In addition, in this embodiment, the shortest path problem (minimization of the moving distance between subgraphs) in the case of passing through each subgraph SG n once via subgraphs SG _n that are connected to each other is solved using an asymmetric cyclic method. Solve it as a salesman problem. Note that the traveling salesman problem is one of the combinatorial optimization problems.

FIG. 18 is a flowchart of calculation processing for solving the shortest route problem for the base vehicle 2 between subgraphs according to the present embodiment. The second movement route determination unit 423 mathematically models the above problem as an asymmetric traveling salesman problem, and executes calculation processing as shown in FIG. Thereby, the second movement route determination unit 423 determines the shortest movement route of the base vehicle 2, which passes through each of the subgraph data SGD _n for which the movement route in the graph has been optimized. That is, the second movement route determination unit 423 determines a second movement route that includes the shortest movement route of the base vehicle 2 that passes through all the subgraph data SGD _n once.

As shown in FIG. 18, the second movement route determining unit 423 determines the weight of the edge connecting each subgraph in the new graph data NGD configured by a plurality of nodes ND _n corresponding to each subgraph SG _n . (Step S61). The problem of determining edge weights (edge weighting problem) can be mathematically modeled as a mixed integer quadratic programming problem. Therefore, the second movement route determining unit 423 calculates using the following equations [15] to [16] (objective function and formulated constraints), which are mathematical models of the mixed integer quadratic programming problem, Solve the edge weighting problem.
(objective function)

(formulated constraints)

(variable list)
Q: A matrix representing the weight of each edge connecting each subgraph, and Q _AB :=Q(A,B) is the weight of a directed edge indicating the direction of movement from a node of subgraph A to a node of subgraph B. is;
d _ij : Length of the shortest path between node i and node j of the graph representing the road path _PR of the road; and x _ij ∈{0,1}: From node i of subgraph A to node j of subgraph B This is a decision variable that indicates whether or not to move.

In FIG. 17, the takeoff position (flight start position) of the drone 3 on the divided waterway path DP _C of the waterway C expressed by the subgraph data SGD is indicated by "○", and the landing position is indicated by "x". ing. In this case, the takeoff position (flight end position) of the drone 3 corresponds to the start position of the inspection work of the divided waterway path _DPC , and corresponds to the start node of the corresponding subgraph. Further, the landing position of the drone 3 corresponds to the end position of the inspection work of the divided waterway path _DPC , and corresponds to the end node of the corresponding subgraph. Therefore, the second movement route determination unit 423 identifies the shortest edge among all edges connecting the end node of subgraph A and the start node of subgraph B, and calculates Q _AB . Such a method of calculating Q _AB can be formulated as an objective function of a mathematical model, as in the above equation [15]. Note that the takeoff position (flight start position) and landing position (flight end position) of the drone 3 are determined by the first movement route determination unit 422. The takeoff position and landing position of the drone 3 are specified based on the movement start node and movement end node on the determined movement path at the time when the movement path of the drone 3 with respect to the subgraph data SGD is determined. Therefore, the first movement route determination unit 422 sends the takeoff position and landing position of the drone 3 corresponding to the movement start node and movement end node on the determined movement route to the second movement route determination unit 423 as parameter values for calculation. hand over.

(constraint list)
Constraint [16]: All ending nodes of subgraph A need to be reached only once, and all starting nodes of subgraph B need to be reached only once.

The second movement route determination unit 423 uses the above formulas [15] to [16] to determine the weight Q of the edge connecting each subgraph in the new graph data NGD, and then passes through a plurality of subgraphs in the shortest possible time. The movement route (minimum distance closed path) of the base vehicle 2 is determined (step S62).

As described above, the shortest path problem for the base vehicle 2 between subgraphs (the problem of minimizing the travel distance between subgraphs) can be mathematically modeled as an asymmetric traveling salesman problem. Therefore, the second movement route determination unit 423 calculates the above-mentioned asymmetric traveling salesman problem using the following equations [17] to [19] (objective function and formulated constraints), and calculates the subgraph. Solve the problem of minimizing the distance traveled between.
(objective function)

(formulated constraints)

(variable list)
x _st ∈{0,1}: is a decision variable representing whether subgraph s is visited by the base vehicle at time t.

(constraint list)
Constraint [18]: For example, the base vehicle must move from the company to the inspection work site and return to the company again after the work is completed; and Constraint [19]: Each subgraph can only be visited once.

The second movement route determining unit 422 calculates the mathematical model of the asymmetric traveling salesman problem based on the above equations [17] to [19], and determines the movement path between the subgraphs at the time when the problem is solved. is determined as the shortest movement route (minimum distance closed path) of the base vehicle 2.

As described above, in the management device 4 according to the present embodiment, the subgraph generation unit 421 performs an inspection to determine whether the drone 3 can continue to fly without replacing the battery 35 based on the possible continuous flight distance or continuous flight time of the drone 3. Subgraph data SGD corresponding to the section is generated. The first movement route determination unit 422 determines a first movement route including each of the shortest movement routes of the base vehicle 2 and the drone 3 for one subgraph data SGD based on the communicable area CR between the base vehicle 2 and the drone 3. decide. The second movement route determination unit 423 passes through all the subgraph data SGD _n once each based on the takeoff position and landing position of the drone 3 corresponding to the movement start node and movement end node on the determined first movement path. A second travel route including the shortest travel route for the base vehicle 2 is determined.

The movement planning unit 42 writes each data of the determined first movement route and second movement route into the storage unit 44 in order to reflect the data on the movement plan data 443 of the base vehicle 2 and the drone 3.

[Movement planning function (re-planning function)]
The work interruption determination unit 45 determines whether or not the work in progress should be interrupted. The work interruption determination unit 45 determines that the work should be interrupted, for example, based on the following information. The work interruption determination unit 45 determines that the work in progress should be interrupted, based on the abnormality/failure information received from the drone 3 via the transmission/reception unit 40. Further, the work interruption determination unit 45 determines whether the base vehicle 2 is traveling on the road based on the latest traffic information acquired via the public line as the parameter data 442 stored in a predetermined storage area of the storage unit 44. , it is determined whether a traffic phenomenon (for example, traffic congestion, etc.) that adversely affects the driving of the vehicle is occurring. When it is determined that a traffic phenomenon that adversely affects driving is occurring, the work interruption determination unit 45 determines that the work in progress should be interrupted. The work interruption determination unit 45 also determines, as parameter data 442, weather phenomena that have an adverse effect on the flight of the drone 3 (for example, strong winds) in the area where the drone 3 is flying, based on the latest weather information acquired via public lines. , heavy rain, snowfall, etc.) is occurring. If it is determined that a weather phenomenon that has an adverse effect on flight is occurring, the work interruption determination unit 45 determines that the work in progress should be interrupted. In this way, the work interruption determination unit 45 determines whether or not the work in progress should be interrupted based on information regarding the flight environment and flight state of the drone 3 and information regarding external factors that affect the inspection work. Determine.

The work interruption determination unit 45 acquires position information of the base vehicle 2 and the drone 3 at the time of work interruption when it is determined that the work in progress is in a state where it should be interrupted. When it is determined that the work in progress should be interrupted based on the information received from the drone 3, the work interruption determination unit 45 determines that the information received from the drone 3 includes location information (for example, GPS value, etc.). Since this information is included, this information is used as the position information of the drone 3 at the time of work interruption. On the other hand, when it is determined based on the parameter data 442 that the work in progress should be interrupted, the work interruption determination unit 45 transmits the position information (for example, GPS value) of the base vehicle 2 via the position sensor 8. etc.). Then, the work interruption determination section 45 requests the drone 3 to provide position information via the transmission/reception section 40 . As a result, the work interruption determination unit 45 can acquire position information from the drone 3.

Further, when the work interruption determination unit 45 determines that the work in progress should be interrupted, it acquires status information (for example, remaining capacity, etc.) of the battery 35 mounted on the drone 3 at the time of interruption of the work. Note that the method for acquiring the state information of the battery 35 from the drone 3 is substantially the same as the method for acquiring the position information described above, so the above description will be used and detailed description will be omitted here.

The work interruption determination unit 45 transfers the position information and the state information of the battery 35 acquired in this way to the replanning unit 46, so that the base vehicle can efficiently and appropriately carry out the remaining work after the work interruption. 2 and 3 to optimize (re-plan) their travel routes.

The replanning unit 46 optimizes the travel route for the drone 3 to fly and inspect again after the work is interrupted, on the waterway path _PC optimized in advance planning. The replanning unit 46 optimizes the travel route on which the base vehicle 2 travels to follow the flight of the drone 3 after the suspension of work in the road path _PR of the roads around the waterway C. At this time, the replanning unit 46 deletes from the divided waterway path _DPC the path (first path data) that corresponds to the inspected section that was inspected according to the movement route before the work was interrupted. Then, the replanning unit 46 subjects the path (second path data) corresponding to the uninspected section that is scheduled to be carried out after restarting the work to recalculate the travel route. In other words, the replanning unit 46 is configured to exclude the inspected sections up to the time the work was interrupted from being subject to recalculation of the travel route. In this way, the replanning unit 46 optimizes the movement route of the base vehicle 2 and the drone 3 over the entire uninspected work area, and re-plans the movement of the base vehicle 2 and the drone 3 based on the data of the optimized movement route. Update the plan data 443. As a result, an optimal movement plan for two different types of moving objects is created that enables efficient and appropriate continuous inspection even if work is interrupted.

The replanning unit 46 includes a subgraph regenerating unit 461 and a movement route re-determining unit 462.

The subgraph regeneration unit 461 regenerates (reconfigures) subgraph data SGD representing the uninspected section _ST2 in which the inspection work is scheduled to be performed after restarting the work. The subgraph regeneration unit 461 reads graph data GD included in the movement plan data 443 that was being used before the suspension of work from the storage unit 44 . The subgraph regeneration unit 461 determines which work is currently being performed in the graph data GD based on the position information (position information of the drone 3) acquired from the work interruption determination unit 45 as the work interruption position IP(t) and/or the work interruption time t. The edge ED corresponding to the interrupted inspection interruption section is specified. Based on the position of the edge EG of the inspection interrupted section in the graph, the subgraph regeneration unit 461 generates an edge EG corresponding to the inspected section _ST1 in which the work was performed according to the movement route before the work was interrupted, in the graph data GD. Identify. The subgraph regeneration unit 461 deletes the identified edge EG from the graph data GD. As a result, the inspected section _ST1 can be excluded from the replanning target, and the replanning target can be narrowed down to the uninspected section _ST2 , which is scheduled to be inspected after the work is resumed.

The subgraph regeneration unit 461 regenerates the subgraph data SGD from the graph data GD including only the edge EG corresponding to the uninspected section _ST2 . Note that the method for regenerating the subgraph data SGD is substantially the same as the method for regenerating the subgraph data SGD by the subgraph generation unit 421 described above, so the relevant description above will be used and detailed description will be omitted here. However, the configuration conditions of the subgraph data SGD (for example, the number of edges included in the subgraph, etc.) depend on the range in which continuous flight is possible (distance or time in which continuous flight is possible) according to the capacity of the battery 35. Therefore, based on the state information of the battery 35 (remaining capacity of the battery 35 at the time of work interruption) acquired from the work interruption determination unit 45 and the flight speed of the drone 3, the subgraph regeneration unit 461 determines whether the drone 3 The distance or time that the battery 35 can continue to fly without replacing the battery 35 is recalculated. The subgraph regeneration unit 461 recalculates the configuration condition value of the subgraph data SGD using the recalculated continuous flight distance or continuous flight time as one parameter value. The subgraph regeneration unit 461 regenerates subgraph data SGD based on the calculation result of the configuration condition value of subgraph data SGD. Furthermore, the configuration conditions of the subgraph data SGD also depend on the total number of base vehicles 2 and the total number of drones 3. Therefore, the management device 4 receives, via the input device 5, the total number of base vehicles 2 and the total number of drones 3 to be used when restarting work. The subgraph regeneration unit 461 recalculates the configuration condition value of the subgraph data SGD using the input total numerical value as one parameter value.

The movement route re-determination unit 462 determines each movement route (first movement route) of the base vehicle 2 and the drone 3 with respect to one subgraph data SGD (subgraph based on the uninspected section _ST2 ) generated by the subgraph regeneration unit 461. ). The method for determining the shortest route for the base vehicle 2 and the drone 3 within the subgraph is approximately the same as the method for determining the shortest route for the base vehicle 2 and the drone 3 described above by the first movement route determination unit 422. is omitted. However, the shortest route for the base vehicle 2 and the drone 3 within the subgraph depends on the range in which continuous flight is possible (distance or time during which continuous flight is possible) according to the capacity of the battery 35. Therefore, the moving route re-determining unit 462 uses the distance or time during which continuous flight is possible, which has been recalculated by the subgraph regenerating unit 461, as one parameter value. Further, like the first route determining unit 422, the moving route re-determining unit 462 mathematically models the problem of minimizing the moving distance within a subgraph as a mixed integer quadratic programming problem, and calculates the shortest route. At this time, in addition to the assumed conditions [1] to [3] above, the movement route re-determining unit 462 determines that ``the base vehicle 2 and the drone 3 will be interrupted from the position of the work interruption time t on the shortest route before recalculation.''"Resuming inspection work" is also assumed as condition [4].

[Mobile object control function]
The mobile object control unit 43 controls the flight of the drone 3, which is one of the mobile objects. The mobile object control unit 43 refers to the movement plan data 443 of the drone 3 stored in the storage unit 44, and sends a predetermined control signal (for example, a takeoff instruction, etc.) so that the drone 3 flies according to the set movement plan (flight plan). ) and predetermined flight plan data (for example, the landing position at the end of the work). Control signals and flight plan data are transmitted to the drone 3 via the transmitting/receiving section 40.

[Drone]
The drone 3 includes a transmitting/receiving section 50, a position specifying section 51, a flight control section 52, a driving section 53, a measurement data processing section 54, a detecting section 55, an abnormality determining section 56, etc., and each functional section operates in cooperation with each other. By doing so, the specified function is provided. The transmitting/receiving unit 50 is a function of the wireless communication device 54, and transmits and receives data with other devices other than the drone 3, such as the base vehicle 2.
The position specifying unit 51 is a function of the position sensor 33, and specifies the current position of the drone 3 based on a predetermined coordinate system. The position specifying unit 51 transfers the specified current position to the flight control unit 52 as position information (eg, GPS value, etc.).

The flight control unit 52 is a function of the flight control device 30, and transmits a control signal to the drive unit 53, receives detection information from the detection unit 55, receives position information from the position specifying unit 51, etc. Performs aircraft control of Drone 3 regarding flight. For example, the flight control unit 52 controls takeoff and the like of the drone 3 according to a control signal received from the management device 4 via the transmission/reception unit 50. Further, the flight control unit 52 controls the flight of the drone 3 along the route, the landing at a predetermined position, etc. based on the flight plan data received from the management device 4 via the transmission/reception unit 50. Note that the flight control unit 52 also controls autonomous flight based on the detection signal and the current position.

The drive unit 53 is a function of the flight device 31, and controls the drive unit 31D to behave as instructed, for example, in accordance with a control signal from the flight control unit 52, and rotationally drives the propeller 31P. The measurement data processing section 54 performs predetermined data processing on the detection result input from the detection section 55, that is, the measurement data 444, and transmits the processed data to the management device 4 via the transmission/reception section 50. In this embodiment, an example of the measurement data 444 is a captured image of the inside of the waterway C that is the inspection result. In the case of a captured image, the measurement data processing unit 54 performs predetermined image processing (for example, filter processing for emphasizing image characteristics) on the image. Note that the measurement data processing section 54 may be a function included in the flight control device 50. Further, when performing image processing, it may be realized by an integrated circuit (ASIC: Application Specific Integrated Circuit) for image processing. The detection unit 55 is a function of the measurement device 36, and acquires sensing results from various sensors as detection results, and transfers the acquired detection results to the flight control unit 52 and the measurement data processing unit 54 as detection information. In this embodiment, the imaging device 37 is cited as an example of various sensors.

The abnormality determination unit 56 determines whether an abnormality/failure that makes it impossible for the drone 3 to continue flying has occurred. The abnormality/failure that makes it impossible to continue flight includes, for example, a state where the drone 3 contacts/collides with an object and a part of the flight device 31 is damaged. Whether or not such an abnormality/failure has occurred can be determined based on detection information from the detection unit 55. In this embodiment, the measurement device 36 includes an inertial measurement device. Therefore, the detection unit 55 can obtain measurement results of the flight environment and flight condition as detection information without delay via these sensing devices. In this embodiment, for example, inertial measurement results during normal flight (hereinafter referred to as "normal values") are stored in a predetermined storage area of a storage device (not shown) included in the drone 3. The abnormality determination unit 56 refers to a normal value stored in advance and compares the value of the inertial measurement result obtained as detection information from the detection unit 55 (hereinafter referred to as “measured value”) with the normal value. The abnormality determination unit 56 determines that an abnormality/failure that makes continued flight impossible has occurred when the absolute value of the difference between the measured value and the normal value is greater than a predetermined threshold. Furthermore, the abnormality determination unit 56 may predict the type of abnormality/failure that has occurred based on the type of inertial measurement that is determined to be an abnormal value. The abnormality determination unit 56 transfers the prediction result of the type of abnormality/failure to the flight control unit 52 together with the determination result of the occurrence of an abnormality/failure.

In response to this, the flight control unit 52 performs aircraft control to land the drone 3 in which the abnormality/failure has occurred in the safest possible location. Further, at this timing, the flight control unit 52 acquires position information (for example, GPS value, etc.) from the position specifying unit 51, and acquires current status information (for example, remaining capacity, etc.) from the battery 35. The flight control unit 52 transmits the acquired position information and battery 35 status information, as well as information on the abnormality/failure that has occurred (including the prediction result of the type of abnormality/failure) to the management device 4 via the transmitting/receiving unit 50. .

As described above, in the drone 3 according to the present embodiment, the flight control unit 52 causes the drone 3 to fly according to the movement plan transmitted from the management device 4. Thereby, the drone 3 flies as shown in FIGS. 6 and 7 and inspects the inside of the waterway C. Further, the drone 3 detects an abnormality/failure that occurs during flight, notifies the management device 4 of this without delay, and lands at the safest possible location.

[Mobile object management processing (pre-planning processing)]
FIG. 19 is a flowchart showing the management process (pre-planning process) for the base vehicle 2 and the drone 3 according to this embodiment. The process shown in FIG. 19 is executed by the management device 4 of the management system 1. This process is mainly divided into graph generation processing by the graph generation unit 41 (steps S71 to S73) and movement planning processing by the movement planning unit 42 (steps S74 to S77).

[Graph generation process]
The management device 4 reads map data 441 including the waterway C to be inspected and roads around it (step S71). The management device 4 extracts from the map data 441 a waterway path PC representing the entire waterway _C to be inspected and a road path _PR representing roads around the waterway C, and extracts each of the waterway path _PC and the road path _PR . Graph data GD corresponding to is generated (step S72). The management device 4 divides each edge EG constituting the graph data GD into equal intervals and reconstructs the graph data GD (step S73). As a result, in this process, for example, the reconstructed graph data RGD shown in FIG. 12C is generated.

[Movement planning process]
The management device 4 reads the parameter data 442 to be referred to in subsequent processing (step S74). At this time, the management device 4 determines the total number of base vehicles 2 used for inspection work, the total number of drones 3, the distance or time that the drone 3 can continue to fly without replacing the battery 35, the flight speed of the drone 3, and the base vehicle 2. and obtains various parameter values such as the communicable area CR between the drone 3 and the drone 3.

Based on the acquired parameter values, the management device 4 generates subgraph data SGD representing an inspection section in which the drone 3 can perform inspection work without replacing the battery 35 (step S75). At this time, the management device 4 divides each edge EG _n at equal intervals and generates a plurality of subgraph data SGD _n from the reconstructed graph data RGD. Therefore, the number of generated subgraph data SGD is such that the battery 35 of the drone 3 is replaced and the time required to travel between each divided waterway path DP _Cn of the waterway C is minimized in the movement route of the entire inspection work of the waterway C. and the number of edges need to be determined.

Therefore, in this embodiment, the problem of minimizing the number of generated subgraphs is mathematically modeled as a mixed integer quadratic programming problem (MIQP) to solve the problem. The management device 4 performs predetermined calculations using the objective function of the mathematical model and the formulated constraints, and solves the problem of minimizing the number of generated subgraphs. Note that the parameter values used at this time are the total number of drones 3, the distance or time that can be continuously flown without replacing the battery 35, and the flight speed. As a result, in this process, the battery 35 of the drone 3 is replaced, and the optimal number of subgraph data SGD to be generated and the number of edges that minimize the time to travel between each divided waterway path DP _Cn of the waterway C are determined. Ru.

The management device 4 determines the shortest movement route of the drone 3 within the generated subgraph (step S76). That is, the management device 4 determines the optimal movement route for the drone 3 to perform inspection work within the divided waterway path _DPC of the waterway C corresponding to the generated subgraph data SGD. Specifically, the management device 4 determines the movement route of the drone 3 that passes through a plurality of nodes ND _n and a plurality of edges EG _n in one subgraph data SGD in the shortest possible time. Furthermore, the management device 4 determines, with respect to the movement path of the drone 3, the movement path of the base vehicle 2 located at a distance that allows communication with the moving drone 3 via the communication device 7.

Therefore, in this embodiment, the shortest path problem (minimization problem of the movement distance within the subgraph) of the base vehicle 2 and the drone 3 for the subgraph data SGD is mathematically modeled as a mixed integer quadratic programming problem (MIQP) to solve the problem. . Note that the parameter values used at this time are the total number of base vehicles 2 and each value of the communicable area (communicable distance) CR between the base vehicle 2 and the drone 3. As a result, in this process, within the communicable region CR between the drone 3 and the base vehicle 2, a first movement route including the shortest movement paths of the base vehicle 2 and the drone 3 for one subgraph data SGD is determined. Ru.

The management device 4 determines the shortest travel route of the base vehicle 2 between the subgraphs in which the travel route within the graph has been optimized (step S77). That is, the management device 4 determines the shortest travel route for the base vehicle 2 to pass through all the divided waterway paths DP _Cn .

Therefore, in this embodiment, the shortest path problem (minimization of the moving distance between subgraphs) in the case of passing through each subgraph SG n once through subgraphs SG _n that are connected to each other is solved by an asymmetric traveling salesman. The problem is solved by mathematically modeling it as a mixed integer quadratic programming problem (MIQP). Note that the parameter values used at this time are the values of the takeoff position and landing position of the drone 3 corresponding to the movement start node and movement end node on the first movement route. As a result, in this process, a second travel route including the shortest travel route of the base vehicle 2 that passes through all the subgraph data SGD _n once is determined.

The management device 4 outputs each data of the determined first movement route and second movement route to be reflected in the movement plan data 443 of the base vehicle 2 and the drone 3 (step S78).

[Mobile object management processing (replanning processing)]
FIG. 20 is a flowchart showing the management process (replanning process) of the base vehicle 2 and the drone 3 according to this embodiment. The process shown in FIG. 20 is executed by the management device 4 of the management system 1.

The management device 4 determines whether the work in progress should be interrupted (step S81). If it is determined that the work should be interrupted (step S81: YES), the management device 4 acquires the work interruption position IP(t) of the base vehicle 2 and the drone 3 (step S82). At this time, the management device 4 acquires not only the position information but also the state information of the battery 35 from the drone 3. Note that various information on the drone 3 may be actively transmitted from the drone 3 when an abnormality/failure occurs, regardless of an acquisition request from the management device 4.

The management device 4 excludes the inspected section _ST1 from the replanning process based on the work interruption position IP(t) and/or the work interruption time t of the drone 3 (step S83). The management device 4 reads from the storage unit 44 the graph data GD included in the movement plan data 443 that was being used before the work was interrupted. Based on the work interruption position IP(t) and/or the work interruption time t of the drone 3, the management device 4 selects an inspected section _ST1 in the graph data GD where the work was performed according to the movement route before the work interruption. The corresponding edge EG is identified, and the identified edge EG is deleted from the graph data GD. Thereby, the target of the rescheduling process can be narrowed down to the uninspected section _ST2 where the inspection work is scheduled to be carried out after the work is resumed.

The management device 4 reads the parameter data 442 to be referred to in subsequent processing (step S84). At this time, the management device 4 determines the total number of base vehicles 2 used for restarting inspection work, the total number of drones 3, the distance or time that the drone 3 can continue to fly without replacing the battery 35, the flight speed of the drone 3, and the base. Various parameter values such as the communicable area CR between the vehicle 2 and the drone 3 are acquired. Note that the parameter values stored in the storage unit 44 are not directly applied to the parameters related to the distance or time during which continuous flight is possible. The state of the battery 35 when the work is interrupted has changed from the state before the work was interrupted. The state mentioned here includes, for example, remaining capacity (state of charge; SOC), deterioration state (state of health, capacity maintenance rate; SOH), and the like. Therefore, the management device 4 recalculates the continuous flight distance or time based on the remaining capacity included in the state information of the battery 35 acquired from the drone 3 and the flight speed of the drone 3, and sets the recalculated value as a parameter. Apply as value. Regarding this point, the parameters regarding the total number of base vehicles 2 and the total number of drones 3 are also the same. The available number at the time of restarting the work may have changed from the state before the work was interrupted due to the occurrence of an abnormality/failure. Therefore, the management device 4 receives the total number of base vehicles 2 and the total number of drones 3 to be used when restarting work via the input device 5, and applies the input values as parameter values.

Based on the obtained parameter values, the management device 4 regenerates subgraph data SGD that allows the drone 3 to perform inspection work without replacing the battery ₃₅ , targeting the uninspected section ST2 (step S85). That is, the management device 4 generates subgraph data SGD that allows the drone 3 to perform inspection work without replacing the battery 35 from the graph data GD from which the edge EG corresponding to the inspected section _ST1 has been deleted. This process is substantially the same as the process in step S75 described above with reference to FIG. Therefore, the above-mentioned pertinent description will be incorporated and detailed description here will be omitted. Here, the battery 35 of the drone 3 is replaced in the movement route of the entire uninspected work area of the waterway C, and subgraph data SGD is generated such that the time to travel the uninspected section _ST2 of the waterway C is minimized. The number and number of edges need to be determined. Therefore, the management device 4 mathematically models the problem of minimizing the number of generated subgraphs as a mixed integer quadratic programming problem (MIQP) and solves the problem. The management device 4 performs predetermined calculations using the objective function of the mathematical model and the formulated constraints, and solves the problem of minimizing the number of generated subgraphs. Note that the parameter value related to the continuous flight distance or time used at this time is a value recalculated according to the remaining capacity of the battery 35, as described above. As a result, in this process, the battery 35 of the drone 3 is replaced, and the optimal number of subgraph data SGD to be generated and the number of edges that minimize the time required to travel through the uninspected section _ST2 of the waterway C are determined.

The management device 4 re-determines the shortest movement route of the drone 3 within the regenerated subgraph (within the subgraph due to the uninspected section _ST2 ) (step S86). This process is substantially the same as the process in step S76 described above with reference to FIG. Therefore, the above-mentioned pertinent description will be incorporated and detailed description here will be omitted. The management device 4 determines the optimal movement route for the drone 3 to perform inspection work within the uninspected section _ST2 of the waterway C corresponding to the regenerated subgraph data SGD. Therefore, the management device 4 mathematically models the shortest path problem (minimization problem of the movement distance within the subgraph) of the base vehicle 2 and the drone 3 for the subgraph data SGD as a mixed integer quadratic programming problem (MIQP) and solves the problem. . As a result, in this process, within the communicable area CR between the drone 3 and the base vehicle 2, the first movement route including the shortest movement paths of the base vehicle 2 and the drone 3 for one subgraph data SGD is re-determined. be done.

The management device 4 outputs the data of the re-determined first movement route in order to reflect it in the movement plan data 443 of the base vehicle 2 and the drone 3 (step S87).

[effect]
As detailed above, the management system 1 according to the present embodiment provides the following effects. The management system 1 divides the waterway path PC representing the entire waterway _C to be inspected according to the range in which the drone 3 can continue to fly without replacing the battery 35. As a result, a plurality of divided waterway paths DP _Cn (inspection sections corresponding to the distance or time that can be continuously flown among all inspection targets) are generated from the waterway path _PC . Furthermore, the management system 1 optimizes each movement route of the base vehicle 2 and the drone 3 for each divided waterway path DP _Cn (divided inspection section) (the shortest moving distance of the base vehicle 2 and the drone 3 in the divided inspection section). or determining the shortest travel time).

As a result, the management system 1 according to the present embodiment can efficiently work while replacing the battery 35 at an appropriate timing in a wide work range that exceeds the continuous flight range of the drone 3 based on the capacity of the onboard battery 35. can be carried out. Furthermore, in the management system 1, the drone 3 can perform inspection work in the shortest distance (or shortest time) in divided inspection sections where continuous flight is possible without replacing the battery 35.

Then, the management system 1 according to the present embodiment determines whether or not the work in progress should be interrupted. When the management system 1 determines that the state should be interrupted, the management system 1 determines the entire movement route of the inspection work in the base vehicle 2 and the drone 3 based on the work interruption position IP(t) and/or the work interruption time t of the drone 3. Optimize again. More specifically, the management system 1 determines, based on the work interruption position IP(t) and/or the work interruption time t, the path corresponding to the inspected section _ST1 in which the work was performed according to the movement route before the work interruption. is deleted from the waterway path _PC . As a result, the targets to be recalculated are narrowed down to the remaining uninspected section _ST2 excluding the inspected section _ST1 . The management system 1 recalculates and determines the optimal movement route for each of the base vehicle ₂ and the drone 3 for the narrowed-down uninspected section ST2.

As a result, in the management system 1 according to the present embodiment, even if work is interrupted for some reason, the shortest travel route for the uninspected section _ST2 is recalculated at the work site, and appropriate inspections can be continued efficiently. Can be implemented.

Furthermore, the management system 1 according to the present embodiment recalculates the distance or time that the drone 3 can continue to fly without replacing the battery 35 based on the remaining capacity of the battery 35 at the time of work interruption. The management system 1 uses the recalculated continuous flight distance or continuous flight time as one of the decision parameters for the regeneration process of the subgraph data SGD. Thereby, the management system 1 can recalculate the shortest travel route of the drone 3 for the uninspected section ST ₂ based on the distance or time that the drone 3 can continue to fly according to the remaining capacity of the battery 35 after resuming work. That is, in the management system 1, the drone 3 can carry out the inspection work in the shortest distance (or in the shortest time) in the uninspected section _ST2 where the drone 3 can continue flying without replacing the battery 35 even after the work is restarted.

Furthermore, the management system 1 according to the present embodiment recalculates the shortest travel route of the drone 3 for the uninspected section ST ₂ based on the number of drones 3 that can be used at the time of resuming work. Thereby, in the management system 1, even if an abnormality/failure occurs and the number of drones that can be used is reduced, it is possible to continue to carry out efficient and appropriate inspections using the drones 3 that can be used when work is resumed.

Furthermore, the management system 1 according to the present embodiment takes into consideration the communication area CR between the drone 3 and the base vehicle 2 and optimizes the movement route of the base vehicle 2. As a result, the management system 1 allows the base vehicle 2 to travel to the battery 35 replacement point while receiving inspection result data from the moving drone 3 in divided inspection sections where continuous flight is possible without replacing the battery 35. I am made to do so.

Furthermore, the management system 1 according to the present embodiment optimizes the movement route of the base vehicle 2 that connects each divided waterway path DP _Cn whose movement route within the path has been optimized (each divided inspection section is inspected once). (determination of the shortest travel route for the base vehicle 2 to pass). Thereby, the management system 1 can efficiently drive the base vehicle 2 carrying the drone 3 after replacing the battery 35 between each divided inspection section.

As described above, in the management system 1 according to the present embodiment, in work using the drone 3 that requires replacement of the electric battery 35 that supplies driving power, the battery 35 can be replaced at an appropriate timing and the work can be carried out efficiently. Can be implemented. In other words, the management system 1 allows optimal work to be carried out using the drone 3 even in a wide work range that exceeds the range in which the drone 3 can continue to fly. Furthermore, even if the work is interrupted, the management system 1 can continue to carry out efficient and appropriate inspections. That is, in the management system 1, even if the work is interrupted for some reason, the shortest travel route for the uninspected section _ST2 is recalculated at the work site, and appropriate inspections can be continued efficiently.

[Modified example]
FIG. 21 is a functional block diagram of the management device 4a according to the first modification. As shown in FIG. 21, a first device 11 different from the management device 4a may include the graph processing unit 41. In this case, the map data 441 is stored, for example, in a predetermined storage area of a storage unit (not shown) included in the first device 11. Further, the first device 11 transmits graph data GD generated by the graph processing unit 41 to the management device 4. As a result, the management device 4 a receives the graph data GD via the transmitting/receiving section 40 and stores it in the storage section 44 . The management device 4a is configured such that the movement planning unit 42 reads the graph data GD from the storage unit 44 when executing the movement planning process. Note that the first device 11 according to this modification may be the computer system 100 or may be an application-specific integrated circuit (ASIC).

FIG. 22 is a functional block diagram of the management device 4b according to the second modification. As shown in FIG. 22, the second device 12, which is different from the management device 4b, may include the mobile object control section 43. In this case, the second device 12 transmits a request to obtain movement plan data 443 to the management device 4b. When the management device 4 receives the acquisition request via the transmitting/receiving unit 40 , it transmits the movement plan data 443 planned by the movement planning unit 42 and stored in the storage unit 44 to the second device 12 via the transmitting/receiving unit 40 . do. As a result, the received movement plan data 443 is stored, for example, in a predetermined storage area of a storage unit (not shown) included in the second device 12. The second device 12 is configured such that the moving object control section 43 reads the movement plan data 443 from the storage section when executing the moving object control process. Note that the second device 12 according to this modification may be the computer system 100 or may be an application-specific integrated circuit (ASIC).

In the embodiment described above, the problem of minimizing the number of generated subgraphs, the problem of minimizing the movement distance within a subgraph, etc. is mathematically modeled as a mathematical optimization problem. Furthermore, the problem of minimizing the moving distance between subgraphs is mathematically modeled as a combinatorial optimization problem. These mathematical models are not limited to the mixed integer quadratic programming problem and the asymmetric traveling salesman problem described in the above embodiments. The mathematical model may be determined by identifying an appropriate solution method according to the definition of the problem.

In the above embodiment, the base vehicle 2 is driven by the driving operation of the driver Dr, but the technology of the present disclosure is not limited thereto. For example, the base vehicle 2 may be configured to autonomously travel based on sensing data around the vehicle and movement plan data 443. Further, the base vehicle 2 may be configured to travel by remote control based on sensing data around the vehicle and movement plan data 443.

The management system 1 according to the embodiment described above is not limited to a configuration in which one base vehicle 2 and one drone 3 are used for inspection work. A configuration using a plurality of base vehicles 2 and a plurality of drones 3 may be used. In the embodiment described above, the total number of base vehicles 2 and the total number of drones 3 used for inspection work are held as parameter data 442, and the movement planning unit 42 refers to these values and determines the shortest time for each of the base vehicles 2 and drones 3. A configuration for determining a travel route is shown.

In the embodiment described above, the vehicle 2 and the unmanned aerial vehicle 3 are cited as examples of two different moving bodies. The vehicle 2 here includes not only four-wheeled vehicles but also two-wheeled, three-wheeled vehicles, and the like. Further, the unmanned aerial vehicle 3 includes not only a multicopter (rotor-wing aircraft) but also a fixed-wing aircraft, an airship-type flying object, and the like. Examples of mobile objects include not only vehicles 2 and unmanned aerial vehicles 3 that can move on the ground and in the sky, but also ships and small submarines that can move on water and underwater. It is mentioned as. Note that it is desirable to appropriately select these moving bodies depending on the work to be used.

In the above embodiment, an example was given in which the mobile body management system 1 is applied to the inspection work of the waterway C, but the technology of the present disclosure is not limited to this. The technology of the present disclosure can be applied to any work that can be performed using at least two moving bodies. Furthermore, the technology of the present disclosure aims to shorten work time (improve work efficiency) and is effective as an alternative to manual work.

1: Management system, 2: Base vehicle (vehicle), 3: Drone (unmanned aircraft), 4: Management device, 5: Input device, 6: Output device, 7: Communication device, 8: Position sensor, 9: Wireless communication Machine;
40: Transmission/reception section;
41: Graph processing unit, 411: Graph generation unit, 412: Edge division unit;
42: movement planning unit, 421: subgraph generation unit, 422: first route determination unit, 423: second route determination unit;
43: Mobile body control unit;
44: Storage unit, 441: Map data, 442: Parameter data, 443: Movement plan data, 444: Measurement data;
45: Work interruption determination unit;
46: replanning unit, 461: subgraph regeneration unit, 462: movement route redetermination unit;
50: Transmission/reception section;
51: Position identification section, 52: Flight control section, 53: Drive section, 54: Measurement data processing section, 55: Detection section, 56: Abnormality determination section;
100: Computer system, 101: Processor, 102: Memory, 103: Storage, 104: Interface;
GD: graph data, SGD: subgraph data;
ND: node, EG: edge.

Claims (12)

The inspection work of a structure whose inspection range is wide, using a first moving body that performs the inspection work and a second moving body that receives and records result data of the inspection work from the first moving body. A mobile object management system for implementing
In order for the first movable body to perform all inspection work on the structure, the shortest travel route of the first movable body is calculated for all inspection sections, and the a planning unit that plans a movement route for the first moving body;
a determination unit that determines whether or not to interrupt the inspection work;
When it is determined that the inspection work is to be interrupted, the shortest movement of the first moving body is performed for the remaining uninspected sections excluding the inspected sections that were carried out before the inspection work was interrupted in all the inspection sections. a replanning unit that recalculates the route and replans the movement route of the first moving body after restarting the inspection work;
a control unit that controls driving of the first moving body so that the first moving body moves through the uninspected area and performs the inspection work according to the replanned movement route ;
The planning department is
Calculating the shortest travel route of the second movable body for the second movable body to move between the plurality of partial inspection sections, and planning the travel route of the second movable body when performing the inspection work. ,
When using a plurality of second moving bodies when restarting the inspection work,
The replanning department is
A mobile body management system that recalculates the shortest travel route of each of the plurality of second mobile bodies based on the total number of the second mobile bodies that can be used when the inspection work is restarted. The replanning department is
The first path data corresponding to the inspected section is deleted from the path data indicating the movement route of the first mobile object in all the inspection sections, and the uninspected section is scheduled to be carried out after the inspection work is resumed. The mobile body management system according to claim 1, wherein the shortest travel route of the first mobile body is recalculated using second path data corresponding to the second path data. The replanning department is
The inspected section and the uninspected section among all the inspection sections are specified based on the position of the first moving body when the inspection work was interrupted and/or the time when the inspection work was interrupted. A management system for a mobile body according to claim 1 or 2. The first mobile body includes a battery as a power source for supplying power to the drive device,
The replanning department is
The movable body according to any one of claims 1 to 3, wherein the shortest travel route of the first movable body is recalculated based on status information indicating the state of the battery when the inspection work is interrupted. management system. The replanning department is
Based on the remaining capacity of the battery included in the status information, calculate the distance or time that the first mobile object can continue to move without replacing the battery after restarting the inspection work, and calculate the calculated travel distance, or The mobile body management system according to claim 4, wherein the shortest travel route of the first mobile body is recalculated based on the calculated travel time and travel speed. The replanning department is
Based on the communicable distance between the first mobile body and the second mobile body, the movement position of the first mobile body becomes a position where communication is possible with the second mobile body in the uninspected section. The mobile body management system according to any one of claims 1 to 5, wherein the mobile body management system recalculates the shortest travel route of the mobile body. When using a plurality of first moving bodies when restarting the inspection work,
The replanning department is
7. The shortest travel route of each of the plurality of first moving bodies is recalculated based on the total number of the first moving bodies available when the inspection work is resumed. Mobile management system. The determination unit includes:
According to any one of claims 1 to 7, it is determined whether or not to interrupt the inspection work based on information regarding the occurrence of an abnormality or failure that prevents continued movement in the first moving body. mobile management system. The determination unit includes:
Any one of claims 1 to 8, wherein whether or not to interrupt the inspection work is determined based on information regarding the occurrence of a phenomenon that adversely affects movement of the first moving body or the second moving body. A mobile object management system described in . The first mobile body is a flying body, the second mobile body is a vehicle,
The determination unit includes:
Determining whether to interrupt the inspection work based on traffic information regarding traffic congestion on the road on which the second mobile object is traveling and/or weather information regarding weather phenomena in the area on which the first mobile object is flying. 10. The mobile object management system according to claim 9. The inspection work of a structure having a wide inspection range is performed using a first mobile body that performs the inspection work and a second mobile body that receives and records result data of the inspection work from the first mobile body. A mobile body management device for implementing
In order for the first movable body to perform all inspection work on the structure, the shortest travel route of the first movable body is calculated for all inspection sections, and the a planning unit that plans a movement route for the first moving body;
a determination unit that determines whether or not to interrupt the inspection work;
When it is determined that the inspection work is to be interrupted, the shortest movement of the first moving body is performed for the remaining uninspected sections excluding the inspected sections that were carried out before the inspection work was interrupted in all the inspection sections. a replanning unit that recalculates the route and replans the movement route of the first moving body after restarting the inspection work;
a control unit that controls driving of the first moving body so that the first moving body moves through the uninspected area and performs the inspection work according to the replanned movement route ;
The planning department is
Calculating the shortest travel route of the second movable body for the second movable body to move between the plurality of partial inspection sections, and planning the travel route of the second movable body when performing the inspection work. ,
When using a plurality of second moving bodies when restarting the inspection work,
The replanning department is
A mobile body management device that recalculates the shortest travel route of each of the plurality of second mobile bodies based on the total number of the second mobile bodies that can be used when the inspection work is restarted. The inspection work of a structure whose inspection range is wide, using a first moving body that performs the inspection work and a second moving body that receives and records result data of the inspection work from the first moving body. A mobile object management method for implementing
In order for the first movable body to perform all inspection work on the structure, the shortest travel route of the first movable body is calculated for all inspection sections, and the a step of planning a movement route of the first mobile object;
a step of determining whether or not to suspend the inspection work;
When it is determined that the inspection work is to be interrupted, the shortest movement of the first moving body is performed for the remaining uninspected sections excluding the inspected sections that were carried out before the inspection work was interrupted in all the inspection sections. recalculating the route and replanning the travel route of the first moving body after restarting the inspection work;
controlling the drive of the first moving body so that the first moving body moves through the uninspected area and performs the inspection work according to the replanned movement route ,
The planning steps include:
Calculating the shortest travel route of the second movable body for the second movable body to move between the plurality of partial inspection sections, and planning the travel route of the second movable body when performing the inspection work. ,
When using a plurality of second moving bodies when restarting the inspection work,
The step of rescheduling includes:
A method for managing movable bodies, comprising recalculating the shortest travel route for each of the plurality of second movable bodies based on the total number of second movable bodies that can be used when the inspection work is restarted.

Images