JP2021015457A - Image generation device - Google Patents

Abstract

To provide an image generation device for outputting a traffic scene as a CG image when the traffic scene satisfies a condition of a predetermined traffic scene.SOLUTION: An image generation device includes: a traffic flow simulation section 2 for performing a traffic flow simulation based on predetermined information; a traffic scene information management section 3 for generating traffic objects based on simulation information from the traffic flow simulation section 2, and managing positions and speeds of the traffic objects and a relative position and relative speed or the like between the objects; an extraction condition storage section 4 for storing a condition to extract a specified traffic scene; a traffic scene extraction section 5 for determining whether a traffic scene to be generated based on the traffic objects managed by the traffic scene information management section 3 satisfies a condition stored in the extraction condition storage section 4; and a CG image generation section 6 for outputting a traffic scene determined to satisfy a condition by the traffic scene extraction section 5 as a computer graphic image.SELECTED DRAWING: Figure 1

Description

The present invention relates to an image generator that generates a specific traffic scene image.

As a means for generating a traffic scene image, for example, there is a technique described in Patent Document 1. Patent Document 1 discloses the following techniques. In Patent Document 1, the computer graphics engine decomposes real-world data into visual components and objects in one or more object categories, and incorporates virtual visual components and virtual objects into the virtual world to bring real-world data into the virtual world. It is cloned. Script components control virtual visual components and virtual objects in the virtual world based on a set of real-world data. Synthetic clones of video sequences are generated based on scripts that control virtual visual components and virtual objects. Real-world data is compared to synthetic clones of video sequences, and based on this comparison, the feasibility of communicating conclusions from the virtual world to the real world is assessed.

Japanese Unexamined Patent Publication No. 2017-151973

Patent Document 1 creates an object in a virtual world by inputting an image and sensor data of a traffic scene when a vehicle is running in the real world, and reproduces the traffic scene by computer graphics. Therefore, in order to generate a traffic scene image by computer graphics, an image of a traffic scene when a vehicle is running in the real world and sensor data are required. Even if the video and sensor data when the vehicle is running in the real world are acquired, the target traffic scene is not always obtained immediately, and most of the acquired video and sensor data are often wasted. For this reason, there is a problem that it takes time to construct data for obtaining a traffic scene.

An object of the present invention is to solve the above problems and to provide a video generator that outputs a computer graphics video when a predetermined traffic scene condition is satisfied.

In order to solve the above problem, one of the representative visual devices of the present invention expresses a real traffic scene generated by a traffic flow simulation as a traffic object in a virtual world, and predetermined traffic scene conditions. It is output as a computer graphics image when the condition is satisfied.

More specifically, the present invention relates to a traffic flow simulation unit that performs a traffic flow simulation based on predetermined information in a video generator that outputs a traffic scene as a computer graphic image, and simulation information from the traffic flow simulation unit. A traffic scene information management unit that generates a traffic object based on the above and manages the position, speed, relative position between objects, relative speed, etc. of the traffic object, and an extraction condition that stores conditions for extracting a specific traffic scene. A traffic scene extraction unit that determines whether or not the storage unit and the traffic scene generated based on the traffic object managed by the traffic scene information management unit satisfy the conditions stored in the extraction condition storage unit. It is characterized by including a computer graphics image generation unit that outputs a traffic scene determined to satisfy the conditions by the traffic scene extraction unit as a computer graphic image.

According to the present invention, it is possible to provide a video generator that outputs as a computer graphics video when satisfying a predetermined traffic scene condition.

It is a block diagram which shows the structure of the image generator which concerns on Example 1 of this invention. It is a block diagram which shows the detailed structure of the traffic simulation part 2 which concerns on Example 1 of this invention. It is a block diagram which shows the detailed structure of the traffic scene information management unit 3 in the image generation apparatus which concerns on Example 1 of this invention. It is a figure which shows an example of the road object generation by the road generation part 300 which concerns on Example 1 of this invention. It is a figure which shows an example of the road object generation by the road generation part 300 which concerns on Example 1 of this invention. It is a figure which shows an example of the road object generation by the road generation part 300 which concerns on Example 1 of this invention. It is a figure which shows an example of the road object generation by the road generation part 300 which concerns on Example 1 of this invention. It is a figure which shows an example of the road object generation by the road generation part 300 which concerns on Example 1 of this invention. It is a figure which shows an example of the positioning of the traffic light by the traffic light generation part 301 which concerns on Example 1 of this invention. It is a figure which shows an example of the positioning of the traffic light by the traffic light generation part 301 which concerns on Example 1 of this invention. It is a figure which shows an example of the building object generation by the building generation part 302 which concerns on Example 1 of this invention. It is a figure which shows an example of the building object generation by the building generation part 302 which concerns on Example 1 of this invention. It is a flowchart which shows the operation of the vehicle management part 310 which concerns on Example 1 in this invention. It is a figure which shows an example of the table stored in the extraction condition storage part 4 which concerns on Example 1 of this invention. It is a flowchart which shows the operation of the traffic scene extraction part 5 which concerns on Example 1 of this invention. It is a figure which shows the extraction example of the traffic scene which concerns on Example 1 of this invention. It is a figure which shows the change example of the traffic scene information management part 3 which concerns on Example 1 of this invention. It is a block diagram which shows the structure of the image generator which concerns on Example 2 of this invention. It is a block diagram which shows the detailed structure of the traffic scene information management part 3 which concerns on Example 2 of this invention. It is a block diagram which shows the detailed structure of the traffic flow simulation part 7 which concerns on Example 2 of this invention. It is a figure which shows an example of the table stored in the extraction condition storage part 4 which concerns on Example 2 of this invention.

Hereinafter, examples of the present invention will be described with reference to the accompanying drawings. The moving image in the embodiment of the present invention includes a moving image and a still image.

FIG. 1 is a block diagram showing a configuration of a video generator according to a first embodiment of the present invention. The image generation device 1 is composed of a traffic flow simulation unit 2, a traffic scene information management unit 3, an extraction condition storage unit 4, a traffic scene extraction unit 5, and a CG image generation unit 6 (CG: Computer Graphics).

The traffic flow simulation unit 2 carries out a traffic flow simulation based on predetermined map information, vehicle, pedestrian route information, and the like.

The traffic scene information management unit 3 generates a three-dimensional map object based on the map information (roads, buildings, etc.) and traffic light information in the traffic flow simulation unit 2, and also simulates the traffic flow simulation unit 2 from moment to moment. It receives information (movement information of vehicles and pedestrians), generates it as a three-dimensional traffic object, and manages each position, relative position between traffic objects, relative speed, and the like.

The extraction condition storage unit 4 stores in advance the extraction conditions of a specific traffic scene to be generated by using the video generation device 1.

The traffic scene extraction unit 5 determines whether or not the traffic scene generated based on the traffic object managed by the traffic scene information management unit 3 satisfies the condition stored in the extraction condition storage unit 4.

The CG image generation unit 6 generates and outputs as a computer graphics image a traffic scene determined by the traffic scene extraction unit 5 to satisfy the conditions stored in the extraction condition storage unit 4.

Next, the configuration of the traffic flow simulation unit 2 will be described with reference to FIG. FIG. 2 is a block diagram showing a detailed configuration of the traffic flow simulation unit 2 according to the first embodiment of the present invention.

The traffic flow simulation unit 2 includes a model data storage unit 20 and a simulation unit 21.

The model data storage unit 20 has a map model 201 having map information (roads, intersections, arrangement of trees, positions of buildings, shapes of buildings, etc.) and a signal model having information on traffic lights corresponding to the connection relationship of roads. It includes 202, a vehicle route model 203 having vehicle types and route information of each vehicle, and a pedestrian route model 204 having pedestrian types and route information of each pedestrian. Although not shown, the model data storage unit 20 may be provided with types of public transportation such as railroads and buses, and route information of public transportation.

The simulation unit 21 carries out a traffic flow simulation based on each model (map model 201, signal model 202, vehicle route model 203, pedestrian route model 204) stored in the model data storage unit 20.

The traffic flow simulation unit 2 can also use an existing traffic flow simulator. In this case, each model stored in the model data storage unit 20 can be easily stored by utilizing the function of the simulator. Can be created. In addition, these models can be replaced according to the purpose of use of the user of the video generation device 1. Further, the model data storage unit 20 may be provided with a driver model that expresses the driving characteristics and its distribution of a driver (not shown), a pedestrian model that expresses the walking characteristics of a pedestrian and its distribution, and the like. ..

Next, the configuration of the traffic scene information management unit 3 will be described with reference to FIG. FIG. 3 is a block diagram showing a detailed configuration of the traffic scene information management unit 3 according to the first embodiment of the present invention.

The traffic scene information management unit 3 includes a static object generation unit 30, a dynamic object management unit 31, a target selection unit 32, and a peripheral information calculation unit 33.

The static object generation unit 30 generates a three-dimensional map object based on the map model 201 and the signal model 202 stored in the model data storage unit 20 of the traffic flow simulation unit 2.

A specific configuration of the static object generation unit 30 will be described. The road generation unit 300 generates roads (roadways, sidewalks, intersections, etc.) as three-dimensional objects from the map model and arranges them in the three-dimensional space.

The traffic light generation unit 301 calculates the position of the traffic light from the connection relationship between the lanes of the road described in the signal model 202, and arranges a three-dimensional object that reproduces the traffic light in the three-dimensional space.

The building generation unit 302 generates a three-dimensional object based on the shape of the building described in the map model and arranges it in the three-dimensional space.

Further, although not shown, a three-dimensional object for a static object such as a tree may be generated and arranged in the three-dimensional space.

Next, the dynamic object management unit 31 will be described. The dynamic object management unit 31 receives and manages simulation data at each step by the simulation unit 21 of the traffic flow simulation unit 2.

The specific configuration of the dynamic object management unit 31 will be described. The vehicle management unit 310 receives the position, orientation, and speed of the vehicle as simulation data, and creates, updates, and deletes the three-dimensional object of the corresponding vehicle.

The pedestrian management unit 311 receives the position, orientation, and speed of the pedestrian as simulation data, and creates, updates, and deletes the corresponding three-dimensional object of the pedestrian.

The signal management unit 312 receives the state information of the signal sent based on the connection relationship for each lane of the road as simulation data, and converts it into the lighting information of the signal light in the three-dimensional object of the traffic light generated by the traffic light generation unit 301. To do.

Next, the target selection unit 32 will be described. The target selection unit 32 determines from which of the static objects and dynamic objects managed by the traffic scene information management unit 3 the viewpoint is targeted for video generation. That is, the target selection unit 32 selects a target three-dimensional object that serves as a reference for a viewpoint for generating a traffic scene. For example, among the vehicles / pedestrians simulated by the traffic flow simulation unit 2, the vehicle / pedestrian to be simulated first may be selected, or the vehicle / pedestrian simulated by an input means (not shown) may be selected. can do. In addition, by selecting an arbitrary point on the map, for example, the position of a traffic light on the map, it is possible to generate a scene that simulates a security camera in the city.

In the peripheral information calculation unit 33, other 3D objects existing around the 3D object (target 3D object) that is the viewpoint for image generation selected by the target selection unit 32 and the target 3D object. Calculates the relative relationship with and (relative speed, relative position, etc.) and outputs it as relative relationship information.

Next, a road object generation method by the road generation unit 300 will be described with reference to FIGS. 4 to 8. 4 to 8 are diagrams showing an example of road object generation by the road generation unit 300 according to the first embodiment of the present invention.

The digital road data is represented by a combination of the nodes 3001a and 3001b as shown in FIG. 4 and the link 3002 connecting the nodes 3001a and 3001b. Unique numbers are set for each of the nodes 3001a and 3001b and the link 3002. Nodes 3001a and 3001b are intersections and nodes in the road network representation, and link 3002 is a road section between node 3001a and node 3001b. As shown in FIG. 5, the node holds the coordinates of the external shape point 3003 and the center point 3000 of the node. Further, as shown in FIG. 6, the link holds the node number to be connected and the lane width information 3004.

Using these data, the road generation unit 300 generates the objects shown in FIGS. 7 and 8. As for the node, as shown in FIG. 7, by using a mesh represented by a set 3006 of triangles 3005 composed of a center point 3000 and two external shape points, an object having a complicated shape as shown in the figure can be created. To do. For the link, as shown in FIG. 8, the length of the link 3007, the inclination of the link 3008, and the center point 3009 of the link are calculated from the coordinate information of the node to which the link is connected, and the object has a rectangular shape of 300A. To do. By making these objects thicker, it is possible to express the height of the sidewalk and the like.

Next, a method of positioning the traffic light by the traffic light generator 301 will be described with reference to FIGS. 9 and 10. 9 and 10 are diagrams showing an example of positioning of a traffic light by the traffic light generation unit 301 according to the first embodiment of the present invention. Digital map data includes information about the lighting state of signals for each connection between lanes, but often does not include information about the position of traffic lights. Therefore, positioning can be performed by the method shown in FIGS. 9 and 10. FIG. 9 shows a case where a road in the straight direction 3011 exists with respect to the position of 3010. In this case, the left end point 3012 of the road after crossing the intersection is calculated, and the traffic light is arranged at the point 3013 on the sidewalk side with this point as a reference. FIG. 10 shows a case where the road in the straight direction 3015 does not exist and the road in the right turn direction 3016 exists with respect to the position 3014. In this case, for the position of 3014, the left end point 3017, which is the end of the road, and the left end (upper end in the drawing) point 3018 of the start point of the road after turning right at the intersection are calculated, and the directions are 3015 and 3016, respectively. The traffic light is arranged at the point 301A on the sidewalk side with reference to the intersection 3019 when extended in. In the case of a road traveling on the right side, the same calculation can be performed by considering FIGS. 9 and 10 symmetrically.

Next, the building object generation method by the building generation unit 302 will be described with reference to FIGS. 11 and 12. 11 and 12 are diagrams showing an example of building object generation by the building generation unit 302 according to the first embodiment of the present invention.

The digital map data may have information on the point 3020 corresponding to the two-dimensional top view of the building as shown in FIG. When these points are connected like a line segment 3021, a figure as shown in the figure is obtained. By arranging a wall 3022 having a height as shown in FIG. 12 with respect to this line segment 3021, it can be reproduced as an object of a building. If the map provides a more detailed three-dimensional structure of the building, it may be used, and corresponds to the size of the figure shown by the point 3020 and the line segment 3021 as shown in FIG. , The three-dimensional object of the building may be provided from the outside.

Next, the operation of the vehicle management unit 310 will be described with reference to FIG. FIG. 13 is a flowchart showing the operation of the vehicle management unit 310 according to the first embodiment of the present invention.

In step S101, the list of vehicle ids being simulated is acquired from the traffic flow simulation unit 2 as ListNew, and in step S102, the ListOlds of vehicle ids in the previous simulation step are compared.

From step S103 to step S109, the vehicle ID included in ListNew and ListOld is looped.

In step S104, it is checked whether or not the vehicle Id exists only in ListNew, and if Yes, the process proceeds to step S105, and it is determined that this vehicle Id is the first vehicle to be simulated in this simulation step. A three-dimensional vehicle object corresponding to this vehicle Id is generated. On the other hand, in step S104, it is checked whether or not the vehicle Id exists only in ListNew, and if No, the process proceeds to step S106, and it is checked whether or not the vehicle Id exists in both ListNew and ListOld.

In step S106, in the case of Yes, it is determined that the vehicle Id continues the simulation between the previous simulation step and the current simulation step, and the process proceeds to step S107.

In step S107, the state of the three-dimensional vehicle object (position, orientation, speed if necessary, headlight, brake lamp, and blinker lighting state) corresponding to the vehicle Id generated in step S106 is updated.

In step S106, if No, that is, if the vehicle Id exists only in ListOld, it is determined that the simulation of the corresponding vehicle Id has been completed in the previous simulation step, and the process proceeds to step S108.

In step S108, the three-dimensional vehicle object corresponding to the vehicle Id is deleted. After the above processing is completed for all vehicles (step S109), in step S110, the content of ListOld is updated to the content of ListNew acquired in step S101, and in step S111, the simulation step is performed for the traffic flow simulation unit 2. Instruct to proceed.

The vehicle management unit 310 executes the processes from step S101 to step S111 for each simulation step of the traffic flow simulation unit 2. As a result, each vehicle simulated by the traffic flow simulation unit 2 is managed as a three-dimensional vehicle object. The pedestrian management unit 311 also operates in substantially the same manner by replacing the vehicle Id in FIG. 13 with the pedestrian Id and the vehicle object with the pedestrian object.

Next, the contents of the table stored in the extraction condition storage unit 4 will be described with reference to FIG. FIG. 14 is a diagram showing an example of a table stored in the extraction condition storage unit 4 according to the first embodiment of the present invention.

As shown in FIG. 14, the table of the extraction condition storage unit 4 stores the location, relative speed, relative distance, number of vehicles in the vicinity, etc. for each traffic scene, and each traffic scene satisfies at least one of these items. Prepared as. For example, if you want to acquire all the traffic scenes in a specific place, only the item of "place" in FIG. 14 may be defined, and the other items may be undefined. The location may be a specific location (coordinates or a specific range), or may be an intersection. By using the table of FIG. 14, it is possible to define conditions for dealing with a traffic scene such as a hiyari hat, for example, a sudden stop of a vehicle in front or a pedestrian jumping out onto the roadway.

Next, the operation of the traffic scene extraction unit 5 will be described with reference to FIG. FIG. 15 is a flowchart showing the operation of the traffic scene extraction unit 5 according to the first embodiment of the present invention. In step S201, the three-dimensional object information of the target selected by the target selection unit 32 of the traffic scene information management unit 3 and the three-dimensional object existing around the target three-dimensional object calculated by the peripheral information calculation unit 33. The relative relationship information of the above is acquired, and it is determined whether or not these match the extraction conditions of any of the traffic scenes stored in the extraction condition storage unit 4 in step S202. If Yes in step S202, the traffic scene matched in step S203 and the video output command are output, and the process ends.

If No in step S202, the process ends as it is. As a result, the video output command is output to the CG video generation unit 6 and the computer graphics video is generated only when the conditions of the traffic scene stored in the extraction condition storage unit 4 are met. In addition, since the matching traffic scene is also output, the label corresponding to the computer graphics image can be acquired at the same time.

Next, an example of extracting a traffic scene in the traffic scene extraction unit 5 will be described with reference to FIG. FIG. 16 is a diagram showing an extraction example of a traffic scene according to the first embodiment of the present invention.

In FIG. 16, the vertical axis shows the relative velocity and the horizontal axis shows the time. Line 330 is an example showing a time-series change in the relative speed of the own vehicle when the vehicle in front suddenly decelerates. When the threshold value of the relative speed with respect to the vehicle in front described in the extraction condition storage unit 4 is set as shown by the broken line 331, only the regions 332 and 333 satisfy the condition. Therefore, only the traffic scenes corresponding to these areas are output as computer graphics images.

As described above, according to the first embodiment, it is possible to efficiently acquire the target traffic scene by determining the value to be stored in the extraction condition storage unit 4 according to the scene required by the user. Become.

Further, in the traffic scene information management unit 3, the three-dimensional object selected by the target selection unit 32 may sense the surrounding three-dimensional objects and output the information together.

Next, a modified example of the first embodiment will be described with reference to FIG. FIG. 17 is a diagram showing a modified example of the traffic scene information management unit 3 according to the first embodiment of the present invention.

Blocks with the same function as in FIG. 3 have the same number. In the modified example, in addition to the configuration shown in FIG. 3, a sensor model 34 and a sensor information calculation unit 35 are provided. The sensor model 34 stores a model for simulating the sensing of external sensors such as LIDAR (abbreviation of Light Detection and Ranging or Laser Imaging Detection and Ranging), sonar, and radar. This model may be provided by the vendor of each sensor or the like. The sensor information calculation unit 35 assumes that at least one of the sensors stored in the sensor model 34 is mounted on the three-dimensional target object selected by the target selection unit 32, and uses the sensor model 34 to perform the target. The periphery of the 3D object is sensed in the 3D space. That is, the sensor information calculation unit 35 calculates the sensor information using the sensor model 34. According to the modification, the sensing result can be added to the extraction condition by setting the item related to the sensor in the table stored in the extraction condition storage unit 4, and a more detailed traffic scene can be extracted.

As described above, according to the image generator of the present embodiment, the image and sensor data of the traffic scene when the vehicle is running in the real world becomes unnecessary, and the real traffic scene generated by the traffic flow simulation is converted into a traffic object in the virtual world. It can be output as a computer graphics image only when the conditions of a predetermined traffic scene are satisfied. The image generation device of the present invention can generate images not only from a vehicle but also from the viewpoint of a vehicle such as a pedestrian, public transportation such as a railroad, a construction machine, or a robot.

Next, Example 2 of the present invention will be described with reference to FIGS. 18 to 21. In the second embodiment, the video generator performs a detailed simulation of the behavior of the target vehicle.

FIG. 18 is a block diagram showing a configuration of a video generator according to a second embodiment of the present invention. Those having the same function as those shown in FIG. 1 have the same reference numerals. In the second embodiment, the traffic flow simulation unit 7 has almost the same function as the traffic flow simulation unit 2 of the first embodiment (FIG. 1), except that the information output from the traffic scene information management unit 8 is traffic. It is to be input to the flow simulation unit 7. The traffic scene information management unit 8 includes a vehicle behavior simulation unit 37 inside, and the result is input to the traffic flow simulation unit 7. As a result of the vehicle behavior simulation unit 37 in the traffic scene extracted by the traffic scene extraction unit 5, the vehicle operation amount and the behavior information of the vehicle are output as the vehicle signal by the vehicle signal output unit 9.

Next, the configuration of the traffic scene information management unit 8 will be described with reference to FIG. FIG. 19 is a block diagram showing a detailed configuration of the traffic scene information management unit 3 according to the second embodiment of the present invention.

Those having the same function as those shown as the traffic scene information management unit 3 in FIG. 3 have the same reference numerals. The vehicle model storage unit 36 stores vehicle parameters for performing simulation in the vehicle behavior simulation unit 37, for example, parameters related to vehicle type, behavior of brakes, engine, steering, and the like. This parameter can be changed as needed. The vehicle behavior simulation unit 37 is a dynamic object managed by the dynamic object management unit 31 in the environment generated by the static object generation unit 30 based on the vehicle parameters stored in the vehicle model storage unit 36. Along with performing a behavior simulation in the presence of, a three-dimensional object of the corresponding vehicle is generated, and the position and orientation of the object are controlled according to the behavior simulation result. The vehicle behavior at this time may be made to behave according to a preset maneuver script, or an input device (not shown) may be used.

For example, an operating means such as a steering wheel, an accelerator pedal, or a brake pedal of an automobile is used as a simulated lower external input device so that a user drives a virtual environment composed of three-dimensional objects. As a result of the behavior simulation at this time, the position, orientation, speed, and the like of the vehicle whose behavior is simulated are input to the traffic flow simulation unit 7 as vehicle behavior data. The peripheral information calculation unit 38 provides information on operations such as pedal operation and steering operation to the vehicle by the vehicle behavior simulation unit 37, information on acceleration, behavior such as each acceleration, and information on relative speed and relative position of peripheral vehicles. Calculate.

Next, the configuration of the traffic flow simulation unit 7 will be described with reference to FIG. FIG. 20 is a block diagram showing a detailed configuration of the traffic flow simulation unit 7 according to the second embodiment of the present invention.

Those having the same function as those shown as the traffic flow simulation unit 2 in FIG. 2 have the same reference numerals. The simulation unit 21 simulates the vehicle and pedestrian represented by the vehicle route model 203 and the pedestrian route model 204 in the environment composed of the map model 201 and the signal model 202 stored in the model data storage unit 20. At the same time, a vehicle simulated by the vehicle behavior simulation unit 37 of the traffic scene information management unit 8 is added to the simulation space, and other vehicles and pedestrians are added based on the vehicle behavior input from the traffic scene information management unit 8. Simulate the behavior of.

Next, the contents of the table stored in the extraction condition storage unit 4 will be described with reference to FIG. FIG. 21 is a diagram showing an example of a table stored in the extraction condition storage unit 4 according to the second embodiment of the present invention.

In the second embodiment, as shown in FIG. 21, items such as acceleration and angular acceleration of the own vehicle and operation of the brake pedal are added to the table. This makes it possible to extract traffic scenes according to the amount of operation and behavior of the own vehicle.

According to the image generation device of the second embodiment, not only the relative relationship with the objects around the own vehicle but also the operation amount and the behavior for the own vehicle are added to the extraction conditions, so that the traffic scene when a specific behavior or operation is performed. Can be output as a computer graphics image.

The video generator according to the second embodiment can be used, for example, for designing and verifying an abnormality detection algorithm for an automobile. In the real world, it is extremely difficult to perform a test run with some parts of an automobile broken down, so it is extremely difficult to acquire vehicle behavior, operation, and video data in such a broken down state. Is.

By utilizing the image generator of the second embodiment, the behavior of the failed vehicle is simulated by simulating the failure of a specific part of the vehicle with the parameters related to the vehicle stored in the vehicle model storage unit 36 in a real traffic flow. At the same time, it is possible to extract only the scenes that perform a specific operation or behavior and output the vehicle signal and computer graphics image, which can be utilized for the design and verification of the abnormality detection algorithm.

The present invention is not limited to the above-mentioned examples, and includes various modifications. For example, the above-described embodiment has been described in detail in order to explain the present invention in an easy-to-understand manner, and is not necessarily limited to those having all the described configurations. Further, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Further, it is possible to add / delete / replace a part of the configuration of each embodiment with another configuration. Furthermore, each of the above configurations, functions, processing units, processing means and the like may be realized by hardware by designing a part or all of them by, for example, an integrated circuit. Further, each of the above configurations, functions, and the like may be realized by software by the processor interpreting and executing a program that realizes each function. Information such as programs, tables, and files that realize each function can be stored in a memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.

1 ... Video generator, 2 ... Traffic flow simulation unit, 3 ... Traffic scene information management unit, 4 ... Extraction condition storage unit, 5 ... Traffic scene extraction unit, 6 ... CG image generation unit, 7 ... Traffic flow simulation unit, 8 ... Traffic scene information management unit, 20 ... Model data storage unit, 21 ... Simulation unit, 30 ... Static object generation unit, 31 ... Dynamic object management unit, 32 ... Target selection unit, 33 ... Peripheral information calculation unit, 34 ... Sensor model, 35 ... Sensor information calculation unit, 36 ... Vehicle model storage unit, 37 ... Vehicle behavior simulation unit, 38 ... Peripheral information calculation unit, 201 ... Map model, 202 ... Signal model, 203 ... Vehicle route model, 204 ... Walking Person route model, 300 ... Road generation unit, 301 ... Signal generation unit, 302 ... Building generation unit, 310 ... Vehicle management department, 311 ... Pedestrian management department, 312 ... Signal management department

Claims (11)

In a video generator that outputs a traffic scene as a computer graphic video
A traffic flow simulation unit that performs traffic flow simulation based on predetermined information,
A traffic scene information management unit that generates a traffic object based on simulation information from the traffic flow simulation unit and manages the position, speed, relative position between objects, relative speed, etc. of the traffic object.
An extraction condition storage unit that stores conditions for extracting a specific traffic scene,
A traffic scene extraction unit that determines whether or not a traffic scene generated based on the traffic object managed by the traffic scene information management unit satisfies the conditions stored in the extraction condition storage unit.
A computer graphics image generation unit that outputs a traffic scene determined to satisfy the conditions by the traffic scene extraction unit as a computer graphic image, and a computer graphics image generation unit.
A video generator characterized by being equipped with. In claim 1,
The traffic flow simulation unit
At least, a model data storage unit including a map model, a signal model, a vehicle route model, and a pedestrian route model.
A simulation unit that performs a traffic flow simulation based on the information in the model data storage unit,
A video generator characterized by being equipped with. In claim 2,
The map model is
An image generator characterized by having at least one of information on roads and intersections, information on the arrangement of trees, information on the position of buildings, and information on the shape of buildings. In claim 3,
The signal model is a video generator characterized by having information on a traffic light corresponding to a road connection relationship. In claim 4,
The vehicle route model is an image generation device including a vehicle type and route information of the vehicle. In claim 5,
The pedestrian route model is an image generation device including the type of pedestrian and the route information of the pedestrian. In claim 2,
In the model data storage unit
An image generation device characterized by having information on the types of public transportation such as railroads and buses, and the route information of the public transportation. In any one of claims 2 to 7,
The traffic scene information management department
A static object generation unit that generates three-dimensional objects for static objects such as roads, traffic lights, buildings, and trees stored in the model data storage unit and arranges them in a three-dimensional space.
A dynamic object management unit that creates, updates, and deletes corresponding three-dimensional objects according to the vehicle, pedestrian, and signal states in each step by the simulation unit.
A target selection unit that selects a target 3D object that serves as a reference for the viewpoint for generating a traffic scene,
Peripheral information calculation unit that calculates the relative relationship between another 3D object existing around the target 3D object selected by the target selection unit and the target 3D object and outputs the relative relationship information. When,
A video generator characterized by being equipped with. In claim 8.
The extraction condition storage unit is at least as a condition of the traffic scene,
Information about the location and
Information on the relative relationship between the target 3D object and other 3D objects existing around the target 3D object, and
A video generator characterized by being equipped with. In claim 8.
The traffic scene information management department
A sensor model that stores a model that simulates the sensing of an external sensor,
A sensor information calculation unit that senses the periphery of the target 3D object in a 3D space using the sensor model for the target 3D object selected by the target selection unit.
A video generator characterized by being equipped with. In claim 8.
The traffic scene information management department
A vehicle model storage unit that stores parameters related to the vehicle type and behavior of the vehicle,
A vehicle behavior simulation unit that simulates vehicle behavior using parameters stored in the vehicle model storage unit,
A video generator characterized by being equipped with.

Images