CN115344052B - Vehicle path control method and control system based on improved group optimization algorithm - Google Patents
Vehicle path control method and control system based on improved group optimization algorithm Download PDFInfo
- Publication number
- CN115344052B CN115344052B CN202211264397.6A CN202211264397A CN115344052B CN 115344052 B CN115344052 B CN 115344052B CN 202211264397 A CN202211264397 A CN 202211264397A CN 115344052 B CN115344052 B CN 115344052B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- path
- state
- obstacle
- road
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 title claims abstract description 16
- 230000006870 function Effects 0.000 claims abstract description 53
- 230000009471 action Effects 0.000 claims abstract description 17
- 230000004888 barrier function Effects 0.000 claims abstract description 17
- 238000010801 machine learning Methods 0.000 claims abstract description 13
- 238000005070 sampling Methods 0.000 claims abstract description 5
- 238000011156 evaluation Methods 0.000 claims abstract description 4
- 239000002245 particle Substances 0.000 claims description 32
- 238000005265 energy consumption Methods 0.000 claims description 28
- 238000001514 detection method Methods 0.000 claims description 21
- 238000005381 potential energy Methods 0.000 claims description 15
- 230000007613 environmental effect Effects 0.000 claims description 10
- 238000011217 control strategy Methods 0.000 claims description 7
- 238000012876 topography Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 6
- 230000003993 interaction Effects 0.000 claims description 6
- 238000003909 pattern recognition Methods 0.000 claims description 6
- 230000005484 gravity Effects 0.000 claims description 4
- 230000008961 swelling Effects 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 230000000903 blocking effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000002028 premature Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a vehicle path control method and a control system based on an improved group optimization algorithm, wherein the control method comprises the following steps: selecting a vehicle operating mode; generating a global path plan based on an improved group optimization algorithm; generating a sampling space moving to a plurality of track states in a state space according to the electronic map, the barrier prediction result, the current point and the position of the target point, and generating a plurality of control actions corresponding to the plurality of track states; obtaining an expected reward of each control action in the plurality of control actions based on a machine learning algorithm, scoring the path through an evaluation function, and obtaining a track with the highest score as a local optimal path; the control unit automatically controls the vehicle according to different terrains and planned paths, and can realize optimal path planning of the vehicle and effective obstacle avoidance on various terrains and complex roads.
Description
Technical Field
The invention belongs to the field of intelligent vehicle control, and particularly relates to a vehicle path control method and a vehicle path control system based on an improved group optimization algorithm.
Background
In recent years, with the development of an automatic driving technology, an automatic driving vehicle has been developed, and the application of the automatic driving vehicle can not only free a driver, but also reduce energy consumption and reduce traffic accidents. For autonomous vehicles, path planning and control are important research content of autonomous vehicles. However, the existing automatic driving vehicle has insufficient automatic driving degree, and cannot effectively classify the terrain so as to obtain the optimal path planning. The path planning and active obstacle avoidance capabilities are also insufficient, and the obstacle avoidance can not be effectively carried out on various terrains and complex roads. In the path planning process, optimization performance depends on artificial parameter setting, the convergence speed is easy to be slow, or premature convergence falls into local optimization, and the optimal performance cannot be achieved. In the fitness function setting, the factors such as energy consumption, obstacle avoidance capability, smoothness, path length, path blocking degree and the like of the path are not fully considered, so that the optimal path planning cannot be obtained according to the needs of users. In order to solve the above problems, the present invention provides a vehicle path control method and a control system based on an improved group optimization algorithm.
Disclosure of Invention
The invention provides a vehicle path control method and a vehicle path control system based on an improved group optimization algorithm, aiming at solving the problems of insufficient automatic driving degree of the existing vehicle and the problems in the path planning and obstacle avoidance processes.
The technical scheme of the invention is as follows:
the invention provides a vehicle path control method based on an improved group optimization algorithm, which comprises the following steps: step 1: selecting a vehicle operating mode, wherein the vehicle operating mode comprises a manual driving mode and an automatic driving mode; when the vehicle running mode is a manual driving mode, the vehicle servo braking system is switched to a power-assisted mode; applying a braking force according to the pedal stepping stroke; when the vehicle running mode is switched to the automatic driving mode, acquiring environmental information, road marking lines, traffic signs, obstacle data and vehicle state data of a vehicle through a plurality of laser radars, a vision sensor and a vehicle state sensor unit; analyzing the collected environmental information, road marking, traffic sign, obstacle data and vehicle state data of the vehicle through a pattern recognition algorithm, and performing road marking recognition, traffic sign recognition, obstacle recognition and vehicle state recognition;
step 2: matching the identified information with a stored map to obtain a real-time electronic map, and analyzing a road range available for driving; generating a travelable area and an obstacle area, and displaying the travelable area and the obstacle area in real time in a map;
and step 3: generating a global path plan based on an improved group optimization algorithm, and if no dynamic barrier is detected, driving according to the global path; when the dynamic barrier is detected, obtaining a local path plan based on a machine learning algorithm;
and 4, step 4: classifying the road into a horizontal road, a complex scene road section and a fluctuating road section according to a real-time electronic map and a road terrain classification rule; and automatically controlling the vehicle according to different road topography and path plans.
Preferably, in step 1, when the vehicle operation mode is switched to the automatic driving mode, the automatic driving mode extends the secondary directory, and the secondary directory includes 4 control modes, namely, a shortest distance mode and a time modeA shortest mode, a mode with least energy consumption and a specified point operation mode, wherein different automatic driving modes pass through the weight value
And adjusting, wherein the weight is obtained through neural network training.
Preferably, the generating a global path plan based on the improved group optimization algorithm includes: step 3.1, rasterizing the map; each grid is in a drivable state or an obstacle state; initializing a particle population, wherein the particle population comprises a population scale, an initial position, an initial speed and iteration times; generating a plurality of groups of initial path point sets, namely a plurality of particles, according to the starting point and the target point; step 3.2, calculating the particle fitness by using a fitness function; step 3.3, updating the position and the speed of the particles; step 3.4, obtaining an individual optimal value and a global optimal value according to the fitness function; step 3.5 repeat steps 3.2 to 3.4 until the maximum number of iterations is reached; step 3.6, outputting a global optimal solution; step 3.7, taking the output optimal solution as a path point; interpolating the path points to generate a smooth path; wherein the fitness function is:
(ii) a WhereinX k ={
,
,…,
}，X k Is as followskParticles, each particle being a path,
is as followskA first particle ofiThe point of the path is the point of the path,
the constant number is a constant number,
，
in order to be a function of avoiding obstacles,
as a function of path distance;
is a path smoothness function;
is a function of energy consumption;
as a congestion function of the waypoints.
1) An obstacle avoidance function:
；
wherein
The distance of the neighboring waypoint vector to the center of the obstacle,
；
is shown asjThe radius of the individual obstacle;
a swelling factor that is an obstacle; when the temperature is higher than the set temperature
When the number is 1, the path is safe, and the obstacle can be avoided; when the temperature is higher than the set temperature
A value of 0 indicates the presence of an obstacle.
2) Path distance function:
wherein
As is the distance from the starting point to the end point,
is the path length; the shorter the length of the path is the shorter,
the larger.
3) Path smoothness function:
， 
(i=1,2,…,n)；
4) Energy consumption function:
；
；
5) Congestion function:
(ii) a Collection of
Representing points of a path
The number of occurrences; which represents the specific gravity of the waypoint relative to the total waypoint during the T period.
Preferably, officeThe partial path planning includes: predicting obstacles in a road according to the electronic map and the detection information, generating a sampling space moving to a plurality of track states in a state space based on the electronic map, a prediction result, the current point and the position of a target point, and generating a plurality of control actions corresponding to the plurality of track states; obtaining rewards of control actions corresponding to each control parameter based on a machine learning algorithm, scoring the paths through a reward function, and obtaining a path with the largest reward as a local optimal path; the reward is:
；
；
；
(ii) a Wherein,
indicating a state when the luggage van reaches the terminal;
indicating a state when the distance between the luggage van and the obstacle is less than a set threshold value;
representing that the state potential value of the current moment is less than the state of the state potential value of the last moment;
the potential energy value state which can be larger than the previous time state at the current time state is represented;
represents other states;
representing a distance to the target;
is a penalty factor;
a prize value for the current time;
the potential energy value of the current state represents the potential energy between the current state and a target point;
is the observed state of the vehicle at time t;
is in a state
The control strategy of (2).
Preferably, the automatically controlling the vehicle according to different road topography and path plans includes: on a horizontal road, the vehicle can run along the central line of the road and can be controlled according to the speed of the vehicle and the speed limit of the road; and for the complicated road sections and the fluctuating road sections, the speed limit control of the classified roads is realized.
Further, the vehicle path control system based on the improved group optimization algorithm comprises a man-machine interaction module, a detection module, a map construction module, a path planning module and a control module; a user selects a vehicle running mode through a man-machine interaction module, and displays an electronic map and path planning information, wherein the vehicle running mode comprises a manual driving mode and an automatic driving mode; the automatic driving mode can be expanded by a secondary directory, and the secondary directory comprises 4 control modes, namely a shortest distance mode, a shortest time mode, a least energy consumption mode and a specified point operation mode; the detection module collects environmental information, road marking, traffic signs, barrier data and vehicle state data of the vehicle; analyzing the data collected by the detection module through a pattern recognition algorithm, and performing road marking recognition, traffic sign recognition, obstacle recognition and vehicle state recognition; the map construction module matches the information identified by the detection module with a stored map to obtain a real-time electronic map, and analyzes a road range available for driving; generating a drivable area and an obstacle area, and displaying the drivable area and the obstacle area in a map in real time; the path planning module generates a global path plan based on an improved group optimization algorithm, and if no dynamic barrier is detected, the vehicle runs according to the global path; when a dynamic barrier is detected, obtaining a local path plan based on a machine learning algorithm; the control module classifies the road into a horizontal road, a complex scene road section and a fluctuating road section according to the real-time electronic map and the road terrain classification rule; and the vehicle is automatically controlled according to different road terrains and path plans.
Preferably, the detection module comprises a vision sensor for obtaining road marking, traffic sign and obstacle detection data; a plurality of laser radars for obtaining a distance from an obstacle to a vehicle and speed information of the obstacle; a vehicle state sensor for obtaining state data of the vehicle, including a position, a speed, an angular velocity of the vehicle; and the data fusion module is used for fusing data of the vision sensor, the laser radar and the vehicle state sensor.
The invention has the beneficial effects that:
1. when the vehicle runs on a complex road, the terrain can be classified according to different terrains, and classification control is performed according to different terrains; the classification control includes setting restrictions on the vehicle speed, steering, and the like. 2. Manual driving and automatic driving can be realized; when the vehicle running mode is manual driving, the vehicle servo braking system is switched to a power-assisted mode; applying a braking force according to the pedal stepping stroke; when the vehicle running mode is switched to automatic driving, the vehicle servo braking system responds to a control command sent by the controller to control the vehicle; the automatic driving mode can be expanded by a secondary directory, the secondary directory comprises 4 control modes including a shortest distance mode, a shortest time mode, a least energy consumption mode and a specified point operation mode, and different automatic driving modes are adjusted through weights. 3. The global optimal path is obtained through an improved group optimization algorithm, the optimal path under different vehicle running modes can be planned by setting different fitness functions through weights, the fitness function of the particle swarm optimization algorithm is improved, and the global path which is safer, smoother and efficient is generated by setting the fitness function and considering the energy consumption, obstacle avoidance capability, path smoothness, path length and path blocking degree of the path. 4. The local path planning adopts a machine learning algorithm to realize the obstacle avoidance of the dynamic obstacles and the local path planning, avoids falling into local optimum by changing a reward function, and realizes effective dynamic obstacle avoidance.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a flow chart of an improved group optimization algorithm;
FIG. 3 is a schematic view of the safe distance of an obstacle;
fig. 4 is a block diagram of the system of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The first embodiment: as shown in fig. 1, the present invention provides a vehicle path control method based on an improved group optimization algorithm, the method comprising the steps of:
step 1: selecting a vehicle operating mode, wherein the vehicle operating mode comprises a manual driving mode and an automatic driving mode; when the vehicle running mode is the manual driving mode, the vehicle servo braking system is switched to the power-assisted mode; applying a braking force according to the pedal stepping stroke; when the vehicle running mode is switched to the automatic driving mode, acquiring environmental information, road marking lines, traffic signs, obstacle data and vehicle state data of a vehicle through a plurality of laser radars, a vision sensor and a vehicle state sensor unit; analyzing the data collected by the detection module through a pattern recognition algorithm, and performing road marking recognition, traffic sign recognition, obstacle recognition and vehicle state recognition;
step 2: matching the information identified by the detection module with a stored map to obtain a real-time electronic map, and analyzing a road range available for driving; generating a drivable area and an obstacle area, and displaying the drivable area and the obstacle area in a map in real time;
and 3, step 3: generating a global path plan based on an improved group optimization algorithm, and if no dynamic barrier can be detected, driving according to the global path; when the dynamic barrier is detected, obtaining a local path plan based on a machine learning algorithm;
and 4, step 4: classifying the road into a horizontal road, a complex scene road section and a fluctuating road section according to a real-time electronic map and a road terrain classification rule; and automatically controlling the vehicle according to different road topography and path plans.
Further, in step 1, when the vehicle operation mode is switched to the automatic driving mode, the automatic driving mode expands the secondary directory, the secondary directory includes 4 control modes, which are respectively the shortest distance mode, the shortest time mode, the least energy consumption mode and the specified point operation mode, wherein different automatic driving modes pass through the weight
And adjusting, wherein the weight is obtained through neural network training.
Further, as shown in fig. 2, the generating the global path plan based on the improved group optimization algorithm includes: step 3.1, rasterizing the map; each grid is in a drivable state or an obstacle state; initializing a particle population, wherein the particle population comprises a population scale, an initial position, an initial speed and iteration times Maxn; from the starting and target points, sets of initial path points, i.e., particles, are generatedX k ，k=1,2, \8230Maxn, one path per particle includingnOne roadA diameter point; step 3.2, calculating the particle fitness by using a fitness function; step 3.3, updating the position and the speed of the particles; step 3.4, obtaining an individual optimal value and a global optimal value according to the fitness function; step 3.5 repeating steps 3.2 to 3.4 until the maximum number of iterations is reached; step 3.6, outputting a global optimal solution; step 3.7, taking the output optimal solution as a path point; interpolating the path points to generate a smooth path; wherein the fitness function is:
(ii) a WhereinX k ={
,
,…,
}，X k Is as followskParticles, each particle being a path,
is as followskA first particle ofiThe points of the path are taken as the path points,
the constant number is a constant number,
，
in order to realize the function of avoiding the obstacle,
as a function of path distance;
is a path smoothness function;
is an energy consumption function;
as a function of congestion at the waypoint.
1) An obstacle avoidance function:
(ii) a As shown in fig. 3, point O is the origin of the obstacle, and the distance between the path and the obstacle is:
then, then
，
Distance from the adjacent path point vector to the center of the obstacle;
denotes the firstjThe radius of the individual obstacle;
a swelling factor that is an obstacle; when in use
When the number is 1, the path is safe, and the obstacle can be avoided; when in use
A value of 0 indicates the presence of an obstacle.
2) Path distance function:
in which
From a starting point to an end pointThe distance between the first and second electrodes,
is the path length; the shorter the length of the path is the shorter,
the larger.
3) Path smoothness function:
， 
(i=1,2,…,n)；
4) Energy consumption function:
；
；
5) Congestion function:
(ii) a Collection
Representing points of a path
The number of occurrences; which represents the specific gravity of the waypoint relative to the total waypoint over the T period.
A second embodiment, as shown in fig. 4, provides a vehicle path control system based on an improved group optimization algorithm, which can be used for path planning and obstacle avoidance of a vehicle, and can be used for devices such as a cargo loading vehicle and an AGV, and includes a human-computer interaction module, a detection module, a map construction module, a path planning module, and a control module;
a user selects a vehicle running mode through a man-machine interaction module, and displays an electronic map and path planning information, wherein the vehicle running mode comprises a manual driving mode and an automatic driving mode; the automatic driving mode can be expanded by a secondary directory, and the secondary directory comprises 4 control modes, namely a shortest distance mode, a shortest time mode, a least energy consumption mode and a specified point operation mode; the detection module collects environmental information, road marking, traffic signs, barrier data and vehicle state data of the vehicle; analyzing the environmental information, road marking, traffic sign, obstacle data and vehicle state data collected by the detection module through a pattern recognition algorithm, and performing road marking recognition, traffic sign recognition, obstacle recognition and vehicle state recognition; the map construction module matches the information identified by the detection module with a stored map to obtain a real-time electronic map, and analyzes a road range available for driving; generating a travelable area and an obstacle area, and displaying the travelable area and the obstacle area in real time in a map; the route planning module generates a global route plan based on an improved group optimization algorithm, and if the dynamic barrier cannot be detected, the vehicle runs according to the global route; when the dynamic barrier is detected, obtaining a local path plan based on a machine learning algorithm; the control module classifies the road into a horizontal road, a complex scene road section and a fluctuating road section according to a real-time electronic map and a road terrain classification rule; and automatically controlling the vehicle according to different road topography and path plans.
Further, the detection module comprises a vision sensor for obtaining detection data of road marking, traffic sign and obstacle; a plurality of laser radars for obtaining a distance from an obstacle to a vehicle and speed information of the obstacle; a vehicle state sensor for obtaining state data of the vehicle, including a position, a speed, an angular velocity of the vehicle; and the data fusion module is used for fusing data of the vision sensor, the laser radar and the vehicle state sensor. It should be noted that the vehicle state sensor may include GPS, compass, inertial navigation system, speed sensor, and other devices.
Further, the local path planning includes: predicting dynamic obstacles according to an electronic map and detected dynamic obstacle information, generating a sampling space moving to a plurality of track states in a state space based on an environment map, an obstacle prediction result, and positions of a current point and a target point, and generating a plurality of control actions corresponding to the plurality of track states; and obtaining expected rewards of each control action in the plurality of control actions based on a machine learning algorithm, scoring the paths through an evaluation function, and obtaining the track with the highest score as a local optimal path.
Preferably, to avoid the algorithm falling into local optima, a reward is obtained at the current reward plus 0.8 times the reward of the previous state:
；
；
；
(ii) a Wherein,
indicating a state when the luggage van reaches the terminal;
indicating a state when the distance between the luggage van and the obstacle is less than a set threshold value;
representing that the state potential value of the current moment is less than the state of the state potential value of the last moment;
the potential energy value state which can be larger than the previous time state at the current time state is represented;
represents other states;
indicating distance to targetSeparating;
is a penalty factor;
a prize value for the current time;
is the observed state of the vehicle at time t;
is a state
The control strategy of (2);
the potential energy value of the state at the current moment represents the potential energy between the current moment and the target point; the potential energy value of the current moment state is smaller than that of the previous moment state, so that the luggage van arrives at the position close to the target from the position far away from the target, and a reward value is added to the current point; the potential energy value which can be larger than the previous time state in the current time state indicates that the luggage van reaches the position far away from the target from the position close to the target, the reward value is reduced for the current point, the state potential energy is introduced, when the luggage van approaches the target point, a certain reward can be obtained, otherwise, the reward is reduced, and therefore a better convergence effect is achieved.
Further, the machine learning algorithm comprises an Actor network and a Critic network, wherein the Actor network is used for determining a control action corresponding to a path state to form a new motion state; the Critic network is used for determining the reward of the control action based on the given path state; the Actor network observes the state according to the current particlesAnd objectsgSelecting an appropriate control actionaObtaining an expected reward by computing a reward functionrThen, the state is fromsIs transferred tos′Will bes，g，a，r，s′The combination is one tuple X =: (s, g,a,r,s′) And deposit it in an experience replayIn a pool; the expected reward for each action is accumulated to calculate a merit function,
wherein E is the mathematical expectation,
as a cost factor; iterating according to a Bellman equation until strategy parameters converge to be optimal; the bellman equation is described as follows:
Further, the automatically controlling the vehicle according to different road topography and path plans includes: on a horizontal road, the vehicle can run along the central line of the road and can be controlled according to the speed of the vehicle and the speed limit of the road; and for the complicated road sections and the fluctuating road sections, the speed limit control of the classified roads is realized.
The above description is only exemplary of the present invention and should not be taken as limiting, and any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. A vehicle path control method based on an improved group optimization algorithm, characterized in that the method comprises the steps of:
step 1: selecting a vehicle operating mode, wherein the vehicle operating mode comprises a manual driving mode and an automatic driving mode; when the vehicle running mode is the manual driving mode, the vehicle servo braking system is switched to the power-assisted mode; applying a braking force according to the pedal stepping stroke; when the vehicle running mode is switched to the automatic driving mode, acquiring environmental information, road marking lines, traffic signs, obstacle data and vehicle state data of a vehicle through a plurality of laser radars, a vision sensor and a vehicle state sensor unit; analyzing the collected environmental information, road marking, traffic sign, obstacle data and vehicle state data of the vehicle through a pattern recognition algorithm, and performing road marking recognition, traffic sign recognition, obstacle recognition and vehicle state recognition;
and 2, step: matching the identified information with a stored map to obtain a real-time electronic map, and analyzing a road range available for driving; generating a travelable area and an obstacle area, and displaying the travelable area and the obstacle area in real time in a map;
and step 3: generating a global path plan based on an improved group optimization algorithm, and if no dynamic barrier can be detected, driving according to the global path; when the dynamic barrier is detected, obtaining a local path plan based on a machine learning algorithm;
and 4, step 4: classifying the road into a horizontal road, a complex scene road section and a fluctuating road section according to a real-time electronic map and a road terrain classification rule; automatically controlling the vehicle according to different road terrains and path plans;
the generating a global path plan based on the improved group optimization algorithm comprises: step 3.1, rasterizing the map; each grid is in a drivable state or an obstacle state; initializing the particle population, including population size, initial position, initial speedDegree and iteration times; generating a plurality of groups of initial path point sets, namely a plurality of particles, according to the starting point and the target point; step 3.2, calculating the particle fitness by using a fitness function; step 3.3, updating the position and the speed of the particles; step 3.4, obtaining an individual optimal value and a global optimal value according to the fitness function; step 3.5 repeat steps 3.2 to 3.4 until the maximum number of iterations is reached; step 3.6, outputting a global optimal solution; step 3.7, taking the output optimal solution as a path point; interpolating the path points to generate a smooth path; wherein the fitness function is:
(ii) a WhereinX k ={
,
,…,
}，X k Is a firstkThe number of the particles is one,
is a firstkA first particle ofiThe points of the path are taken as the path points,
the constant number is a constant number,
，
in order to realize the function of avoiding the obstacle,
(ii) a Wherein
The distance of the neighboring waypoint vector to the center of the obstacle,
is shown asjThe radius of the individual obstacle;
a swelling factor that is an obstacle;
as a function of the distance of the path,
wherein
As is the distance from the starting point to the end point,
is the path length; the shorter the length of the path is the shorter,
the larger;
as a function of the smoothness of the path,
， 
，i=1,2,…,n，
，
(ii) a Wherein
For angular variation of adjacent paths, in the range 0, pi]；
In order to be a function of the energy consumption,
；
；
(ii) a Wherein
In order to be a point of the path,
respectively, the coordinates of the path points are represented,
is the terrain height of the waypoint;
>1> 
(ii) a Coefficient of passage
Distinguishing different energy consumption of the vehicle when going uphill and downhill; when the vehicle goes uphill, the energy consumption is increased, and when the vehicle goes downhill, the energy consumption is reduced;
as a function of the congestion at the waypoint,
(ii) a Collection
Representing points of a path
The number of occurrences; it represents the specific gravity of the waypoint relative to the total waypoint during the T period;
the local path planning comprises: predicting a dynamic obstacle according to an electronic map and detected dynamic obstacle information, generating a sampling space moving to a plurality of track states in a state space based on an environment map, an obstacle prediction result, a current point and the position of a target point, and generating a plurality of control actions corresponding to the plurality of track states; obtaining an expected reward of each control action in the plurality of control actions based on a machine learning algorithm, scoring the path through an evaluation function, and obtaining a track with the highest score as a local optimal path; the reward is as follows:
；
；
；
(ii) a Wherein,
indicating a state when the vehicle reaches the terminal;
indicating a state when the distance between the vehicle and the obstacle is less than a set threshold;
representing that the state potential value of the current moment is less than the state of the state potential value of the last moment;
the potential energy value state which can be larger than the previous time state at the current time state is represented;
represents other states;
representing a distance to the target;
is a penalty factor;
a prize value for the current time;
the potential energy value of the state at the current moment represents the potential energy between the current moment and the target point;
is in state for control strategy
The prize to be paid down is given,
the observed state of the vehicle at time t;
is in a state
The control strategy of (1).
2. The control method according to claim 1, wherein in step 1, when the vehicle operation mode is switched to the automatic driving, the automatic driving mode extends a secondary directory, the secondary directory includes 4 control modes, which are a shortest distance mode, a shortest time mode, a least energy consumption mode, and a designated point operation mode, respectively, wherein different automatic driving modes pass through the weight value
And adjusting, wherein the weight is obtained through neural network training.
3. The control method of claim 1, wherein automatically controlling the vehicle according to different road topography and path plans comprises: on a horizontal road, the vehicle can run along the central line of the road and can be controlled according to the speed of the vehicle and the speed limit of the road; and for the complicated road sections and the fluctuating road sections, the speed limit control of the classified roads is realized.
4. A vehicle path control system based on an improved group optimization algorithm is characterized by comprising a man-machine interaction module, a detection module, a map construction module, a path planning module and a control module;
a user selects a vehicle running mode through a man-machine interaction module, and displays an electronic map and path planning information, wherein the vehicle running mode comprises a manual driving mode and an automatic driving mode; the automatic driving mode can expand a secondary directory, wherein the secondary directory comprises 4 control modes, namely a shortest distance mode, a shortest time mode, a minimum energy consumption mode and a specified point operation mode;
the detection module collects environmental information, road marking, traffic signs, barrier data and vehicle state data of the vehicle; analyzing the environmental information, road marking, traffic signs, obstacle data and vehicle state data collected by the detection module through a pattern recognition algorithm, and performing road marking recognition, traffic sign recognition, obstacle recognition and vehicle state recognition;
the map construction module matches the information identified by the detection module with a stored map to obtain a real-time electronic map, and analyzes a road range available for driving; generating a drivable area and an obstacle area, and displaying the drivable area and the obstacle area in a map in real time;
the route planning module generates a global route plan based on an improved group optimization algorithm, and if the dynamic barrier cannot be detected, the vehicle runs according to the global route; when the dynamic barrier is detected, obtaining a local path plan based on a machine learning algorithm;
the control module classifies the road into a horizontal road, a complex scene road section and a fluctuating road section according to a real-time electronic map and a road terrain classification rule; automatically controlling the vehicle according to different road terrains and path plans;
the generating a global path plan based on the improved group optimization algorithm comprises: 1) Rasterizing a map; each grid is in a travelable state or an obstacle state; initializing a particle population, wherein the particle population comprises a population scale, an initial position, an initial speed and iteration times; generating a plurality of groups of initial path point sets, namely a plurality of particles, according to the starting point and the target point; 2) Calculating particle fitness using a fitness function; 3) Updating the particle position and velocity; 4) Obtaining an individual optimal value and a global optimal value according to the fitness function; 5) Repeating the steps 3) to 4) until the maximum iteration number is reached; 6) Outputting a global optimal solution; 7) Taking the output optimal solution as a path point; interpolating the path points to generate a smooth path; wherein the fitness function is:
(ii) a WhereinX k ={
,
,…,
}，X k Is as followskThe number of the particles is one,
is a firstkA first particle ofiThe points of the path are taken as the path points,
the constant number is a constant number,
，
in order to realize the function of avoiding the obstacle,
(ii) a Wherein
The distance of the neighboring waypoint vector to the center of the obstacle,
denotes the firstjThe radius of the individual obstacle;
a swelling factor that is an obstacle;
as a function of the distance of the path,
wherein
As is the distance from the starting point to the end point,
is the path length; the shorter the length of the path is the shorter,
the larger;
as a function of the smoothness of the path,
， 
，i=1,2,…,n，
；
(ii) a Wherein
For angular variation of adjacent paths, in the range [0, π]；
In order to be a function of the energy consumption,
；
；
(ii) a Wherein
In order to be a point of the path,
the coordinates of the path points are respectively represented,
is the terrain height of the waypoint;
>1> 
(ii) a Coefficient of passage
Distinguishing different energy consumption of the vehicle when going uphill and downhill; when the vehicle goes uphill, the energy consumption is increased, and when the vehicle goes downhill, the energy consumption is reduced;
as a function of the congestion at the waypoint,
(ii) a Collection
Representing points of a path
The number of occurrences; it represents the specific gravity of the waypoint relative to the total waypoint during the T period;
the local path planning comprises: predicting dynamic obstacles based on the electronic map and the detected dynamic obstacle informationGenerating a sampling space moving to a plurality of track states in a state space according to the environment map, the barrier prediction result, the current point and the position of a target point, and generating a plurality of control actions corresponding to the plurality of track states; obtaining an expected reward of each control action in the plurality of control actions based on a machine learning algorithm, scoring the path through an evaluation function, and obtaining a track with the highest score as a local optimal path; the reward is as follows:
；
；
；
(ii) a Wherein,
indicating a state when the vehicle reaches the terminal;
indicating a state when the distance between the vehicle and the obstacle is less than a set threshold value;
representing that the state potential value of the current moment is less than the state of the state potential value of the last moment;
the potential energy value state which can be larger than the previous time state at the current time state is represented;
represents other states;
representing a distance to the target;
is a penalty factor;
a prize value for the current time;
the potential energy value of the state at the current moment represents the potential energy between the current moment and the target point;
is in state for control strategy
The prize to be paid down is given,
the observed state of the vehicle at time t;
is in a state
The control strategy of (2).
5. The system of claim 4, wherein the detection module comprises a vision sensor for obtaining road marking, traffic sign, and obstacle detection data; a plurality of laser radars for obtaining a distance from an obstacle to a vehicle and speed information of the obstacle; a vehicle state sensor for obtaining state data of the vehicle, including a position, a speed, and an angular velocity of the vehicle; and the data fusion module is used for fusing data of the vision sensor, the laser radar and the vehicle state sensor.
6. The system of claim 4, wherein automatically controlling the vehicle according to the different road topography and path plans comprises: on a horizontal road, the vehicle can run along the central line of the road and can be controlled according to the speed of the vehicle and the speed limit of the road; and for the complicated road sections and the fluctuating road sections, the speed limit control of the classified roads is realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211264397.6A CN115344052B (en) | 2022-10-17 | 2022-10-17 | Vehicle path control method and control system based on improved group optimization algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211264397.6A CN115344052B (en) | 2022-10-17 | 2022-10-17 | Vehicle path control method and control system based on improved group optimization algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115344052A CN115344052A (en) | 2022-11-15 |
| CN115344052B true CN115344052B (en) | 2023-01-13 |
Family
ID=83957158
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211264397.6A Active CN115344052B (en) | 2022-10-17 | 2022-10-17 | Vehicle path control method and control system based on improved group optimization algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115344052B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115100622B (en) * | 2021-12-29 | 2023-09-22 | 中国矿业大学 | Method for detecting driving area of unmanned transportation equipment in deep limited space and automatically avoiding obstacle |
| CN116519005B (en) * | 2023-07-04 | 2023-10-03 | 上海云骥跃动智能科技发展有限公司 | Path planning method and device |
| CN118274865B (en) * | 2024-06-04 | 2024-07-30 | 山东北骏重工有限公司 | Mining vehicle-oriented distributed learning path planning method |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104050390B (en) * | 2014-06-30 | 2017-05-17 | 西南交通大学 | Mobile robot path planning method based on variable-dimension particle swarm membrane algorithm |
| CN109799822A (en) * | 2019-01-30 | 2019-05-24 | 中国石油大学(华东) | Mobile robot global smooth paths planing method |
| CN111738396B (en) * | 2020-06-01 | 2023-09-26 | 北京中安智能信息科技有限公司 | Self-adaptive grid granularity ant colony method applied to submarine path planning |
| CN114440908B (en) * | 2020-10-31 | 2023-07-28 | 华为技术有限公司 | Method and device for planning driving path of vehicle, intelligent vehicle and storage medium |
| CN112286212B (en) * | 2020-12-29 | 2021-03-23 | 北京理工华创电动车技术有限公司 | Vehicle network cooperative energy-saving control method |
| CN113865603B (en) * | 2021-08-30 | 2024-06-07 | 东风柳州汽车有限公司 | Shared unmanned vehicle path planning method, device, equipment and storage medium |
| CN114859905A (en) * | 2022-04-25 | 2022-08-05 | 南京航空航天大学 | Local path planning method based on artificial potential field method and reinforcement learning |
| CN114995425B (en) * | 2022-06-02 | 2024-05-07 | 航天时代(青岛)海洋装备科技发展有限公司 | Unmanned ship autonomous tracking method based on time optimization and energy optimization |
-
2022
- 2022-10-17 CN CN202211264397.6A patent/CN115344052B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115344052A (en) | 2022-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115344052B (en) | Vehicle path control method and control system based on improved group optimization algorithm | |
| US11726477B2 (en) | Methods and systems for trajectory forecasting with recurrent neural networks using inertial behavioral rollout | |
| US11554785B2 (en) | Driving scenario machine learning network and driving environment simulation | |
| CN112888612B (en) | Automated driving vehicle planning | |
| EP3678911B1 (en) | Pedestrian behavior predictions for autonomous vehicles | |
| US12103560B2 (en) | Methods and systems for predicting actions of an object by an autonomous vehicle to determine feasible paths through a conflicted area | |
| US12134409B1 (en) | Methods for cooperative decision-making on lane-changing behaviors of autonomous vehicles based on bayesian game | |
| CN109684702B (en) | Driving risk identification method based on trajectory prediction | |
| CN109643118B (en) | Influencing a function of a vehicle based on function-related information about the environment of the vehicle | |
| CN107886750B (en) | Unmanned automobile control method and system based on beyond-visual-range cooperative cognition | |
| CN114312830B (en) | Intelligent vehicle coupling decision model and method considering dangerous driving conditions | |
| CN116249947A (en) | Predictive motion planning system and method | |
| Shimosaka et al. | Modeling risk anticipation and defensive driving on residential roads with inverse reinforcement learning | |
| CN113255998B (en) | Expressway unmanned vehicle formation method based on multi-agent reinforcement learning | |
| CN113071487B (en) | Automatic driving vehicle control method and device and cloud equipment | |
| WO2022178858A1 (en) | Vehicle driving intention prediction method and apparatus, terminal and storage medium | |
| CN113743469A (en) | Automatic driving decision-making method fusing multi-source data and comprehensive multi-dimensional indexes | |
| Menéndez-Romero et al. | Courtesy behavior for highly automated vehicles on highway interchanges | |
| CN113120003B (en) | Unmanned vehicle motion behavior decision method | |
| CN114715193A (en) | Real-time trajectory planning method and system | |
| US20240174266A1 (en) | Prediction model with variable time steps | |
| CN113743767B (en) | Vehicle dispatching method, system, computer and medium based on time and safety | |
| US20240092398A1 (en) | Trajectory prediction based on a decision tree | |
| CN113415278A (en) | System and process for following a nearest on-path vehicle using a flow of ambient vehicle motion | |
| CN118494468B (en) | Vehicle control method and system based on artificial intelligence |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20221115 Assignee: Jiangsu Tianyi Airport Equipment Maintenance Service Co.,Ltd. Assignor: Jiangsu Tianyi Aviation Industry Co.,Ltd. Contract record no.: X2023980044219 Denomination of invention: Vehicle Path Control Method and Control System Based on Improved Group Optimization Algorithm Granted publication date: 20230113 License type: Common License Record date: 20231024 |