CN115116220A - Unmanned multi-vehicle cooperative control method for loading and unloading scene of mining area - Google Patents

Abstract

The invention discloses an unmanned multi-vehicle cooperative control method for loading and unloading scenes in a mining area, which is characterized in that a multi-vehicle discretization track is obtained based on conflict resolution and a time-space heuristic search algorithm and is used as an initial solution of a cooperative control model, so that the algorithm solving efficiency can be improved; by establishing a nonlinear multi-vehicle cooperative control optimization model, vehicle grouping and conflict priority is taken as a cooperative control strategy, and the load and unload scene characteristics are combined, so that the convergence speed and the solution success rate of the algorithm can be improved; the driving risk estimation is carried out based on the potential energy risk field, so that the complexity of the cooperative control model can be reduced; by designing a multi-objective optimization function and constraint conditions, the multi-vehicle running track planning is carried out under the condition of meeting vehicle motion constraints, the vehicle safety distance can be dynamically adjusted for vehicles with different task functions, the trafficability of group vehicles in a complex and variable environment is improved, and the cooperative loading and unloading operation of a plurality of unmanned vehicles is realized.

Description

Unmanned multi-vehicle cooperative control method for loading and unloading scene of mining area

Technical Field

The invention relates to the technical field of unmanned driving, in particular to an unmanned multi-vehicle cooperative control method for loading and unloading scenes in a mining area.

Background

The mining area loading and unloading scene generally refers to an area allowing vehicles to be loaded and unloaded in the mining area, lane lines and clear road boundaries do not exist under the area, and the loading and unloading transportation vehicles are various. The unmanned multi-vehicle cooperative control method in the loading and unloading scene of the mining area mainly means that a plurality of unmanned vehicles generate safe travelable tracks in the loading and unloading scene according to external environment information, the self state of the vehicles and loading and unloading task requirements. The unmanned vehicle needs to safely travel along the generated trajectory under the constraints of vehicle kinematics.

For an urban road environment, due to the fact that roads are continuous and the road surface is flat, the minimum turning radius of a vehicle is small, and cooperative control among multiple vehicles is easy to achieve. For a complicated and severe loading and unloading scene of a mining area, a control object is a heavy mine truck, and vehicle control has high delay lag, so that how to perform multi-vehicle cooperative control of the heavy truck in the mining area scene is one of the technical difficulties.

In recent years, a path planning method for a robot is provided, vehicle track preliminary planning is completed through heuristic search and a numerical optimization algorithm, urban structured road scenes are mostly involved, application of multi-vehicle cooperation in loading and unloading scenes in mining areas is not considered, the number of vehicles is large in the scenes, the working area is limited, the solving speed of the numerical optimization algorithm is low, and real-time planning of parking paths cannot be guaranteed.

Disclosure of Invention

In view of the above, the invention provides an unmanned multi-vehicle cooperative control method for a loading and unloading scene in a mining area, which is used for solving the problem of multi-vehicle cooperative control in the loading and unloading scene in the mining area with a low structuralization degree and realizing rapid generation of multi-vehicle driving tracks meeting vehicle kinematic constraints in a limited area.

The invention provides an unmanned parking path planning method for loading and unloading scenes in a mining area, which comprises the following steps of:

s1: obtaining starting points P of N vehicles _s,1 ,,P _s,2 ,…,P _s,N And its job task end point P _g,1 ,P _g,2 ,…,P _g,N (ii) a Wherein N is a positive integer;

s2: the starting point of each vehicle is taken as a planning initial point, the operation task end point is taken as a planning target point, and the heuristic search algorithm is adopted to complete the planning of the initial path of each vehicle;

s3: acquiring the number M of path points of the longest path in the initial paths of all vehicles, and judging whether the number of the path points of the remaining vehicles is less than M or not; if yes, the initial path points are supplemented to M, and the supplemented path coordinate points are the operation task end points; time division is carried out on path points on all the initial paths of the vehicles, and a time sequence T is given _d As adjacent waypoint time intervals;

s4: according to the time sequence, judging two vehicles V at the jth path point _i And V _k If j is less than a first threshold value, j is 1, …, M, i is 1,2, …, N, k is 1,2, …, N, i ≠ k; if yes, the running tracks of the two vehicles are respectively re-planned, and safety constraint is added in the planning process

Respectively obtaining the satisfied constraints by using a time heuristic search algorithm

Safety trace trj _i And trj _j Comparing the two tracks, and selecting the optimal track to replace the original track of the corresponding vehicle;

s5, repeating the step S4, and determining two vehicles V at the next route point j ═ j +1 _i And V _k If so, planning is repeated until the planning of the traversable tracks of all vehicles is finished or the planning is finished after the maximum planning times is reached, and the planning is taken as an initial solution;

s6, establishing a nonlinear multi-vehicle cooperative control optimization model comprising a target optimization function, vehicle kinematics constraint, barrier collision constraint and inter-vehicle safe distance constraint, taking all vehicle driving tracks as an initial solution, evaluating vehicle driving risks by adopting a potential energy risk field method, andjudging two vehicles V _i And V _k Whether or not there is a collision at time T, T0, …, (M-1) × T _d 1,2, …, N, k 1,2, …, N, i ≠ k; if yes, adding the moment into the optimization model as an effective safety distance constraint;

and S7, adding all vehicles carrying cargos into the same marshalling by using the nonlinear multi-vehicle cooperative control optimization model to perform track cooperative optimization to obtain the travelable track of each vehicle.

In a possible implementation manner, in the above unmanned multi-vehicle cooperative control method for a loading and unloading scene in a mining area provided by the present invention, in step S2, a heuristic search algorithm is adopted to complete initial path planning for each vehicle, with a starting point of each vehicle as a planning initial point and an end point of a job task as a planning target point, and specifically includes:

generating a drivable path of each vehicle by using a heuristic search algorithm, taking the starting point of each vehicle as a planning initial point, taking the operation task end point as a planning target point, not considering the interactive relation among the vehicles and only considering the obstacles in the scene;

in a possible implementation manner, in the above unmanned multi-vehicle cooperative control method for a loading and unloading scene in a mining area, in step S3, the initial path points are supplemented to M, and the supplemented path coordinate points are job task end points; time division is carried out on route points on all vehicle initial routes, and a time sequence T is given _d As the adjacent path point time interval, the method specifically includes:

make up to M initial path points and add T _d As the time interval between adjacent route points, the initial route driving time of all vehicles is (M-1) × T _d ；

In a possible implementation manner, in the above unmanned multi-vehicle cooperative control method for loading and unloading scenes in a mining area provided by the invention, in the step S4, a time heuristic search algorithm is used to obtain the results respectively satisfying the constraints

Safety trace trj _i And trj _j Is toComparing the two tracks, selecting the optimal track to replace the original track of the corresponding vehicle, and specifically comprising the following steps of:

in the time heuristic algorithm searching process, whether the path point j violates the safety constraint condition or not is judged

If yes, the arrival time of the path point is prolonged until the safety constraint condition is met

Respectively planning vehicles V _i And V _k Travel trajectory, comparative trajectory trj _i And trj _k Length of travel time, if trj _i The running time is more than trj _k Then trj _k Replacement vehicle V _k Original trajectory, otherwise trj _i Replacement vehicle V _i The original trajectory;

in one possible implementation manner, in the above-mentioned unmanned multi-vehicle cooperative control method for loading and unloading scenes in a mining area, in step S5, it is determined that two vehicles V are present at the next route point j ═ j +1 _i And V _k If so, planning repeatedly, and specifically comprising the following steps:

in the repeated planning process, if the vehicle V _i Or V _k Planning before the path point j, and planning by only selecting the path point after the path point j as a starting point in the next planning process;

in a possible implementation manner, in the above unmanned multi-vehicle cooperative control method for a loading and unloading scene in a mine area, in step S7, the method includes that, using a nonlinear multi-vehicle cooperative control optimization model, all vehicles with cargos are added into the same group to perform track cooperative optimization, so as to obtain a traversable track of each vehicle, and further includes:

and selecting the target with the maximum driving risk from the rest vehicles as an object to be optimized, combining the vehicles with the maximum collision with the target, and performing track collaborative optimization as a combination to be optimized.

The unmanned multi-vehicle cooperative control method for the loading and unloading scene of the mining area is provided by the invention, and aims to solve the problems of low solving speed and low success rate of unmanned multi-vehicle cooperative control in the loading and unloading scene of the mining area. The multi-vehicle discretization track is obtained based on the conflict resolution and the time-space heuristic search algorithm and is used as an initial solution of the cooperative control model, so that the algorithm solving efficiency can be improved; by establishing a nonlinear multi-vehicle cooperative control optimization model, vehicle grouping and conflict priority is taken as a cooperative control strategy, and the load and unload scene characteristics are combined, so that the convergence speed and the solution success rate of the algorithm can be improved; the driving risk estimation is carried out based on the potential energy risk field, so that the complexity of the cooperative control model can be reduced; by designing a multi-objective optimization function and constraint conditions, the multi-vehicle running track planning is carried out under the condition of meeting vehicle motion constraints, the vehicle safety distance can be dynamically adjusted for vehicles with different task functions, the trafficability of group vehicles in a complex and variable environment is improved, and the cooperative loading and unloading operation of a plurality of unmanned vehicles is realized.

Drawings

FIG. 1 is a flow chart of an unmanned multi-vehicle cooperative control method for loading and unloading scenes in a mining area according to the present invention;

fig. 2 is a schematic view of the special road scene space segment division and parking path planning in embodiment 1 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 illustrative and are not intended to limit the present invention.

The invention provides a method for planning an unmanned parking path in a special road scene, which comprises the following steps as shown in figure 1:

s1: obtaining starting points P of N vehicles _s,1, ,P _s,2 ,…,P _s,N And its job task end point P _g,1 ,P _g,2 ,…,P _g,N (ii) a Wherein N is a positive integer;

s2: the starting point of each vehicle is taken as a planning initial point, the operation task end point is taken as a planning target point, and the heuristic search algorithm is adopted to complete the planning of the initial path of each vehicle;

s3: acquiring the number M of path points of the longest path in the initial paths of all vehicles, and judging whether the number of the path points of the remaining vehicles is less than M or not; if yes, the initial path points are supplemented to M, and the supplemented path coordinate points are the operation task end points; time division is carried out on path points on all the initial paths of the vehicles, and a time sequence T is given _d As adjacent waypoint time intervals;

s4: according to the time sequence, judging two vehicles V at the jth path point _i And V _k If j is less than a first threshold value, j is 1, …, M, i is 1,2, …, N, k is 1,2, …, N, i ≠ k; if yes, the driving tracks of the two vehicles are planned again respectively, and safety constraints are added in the planning process

Respectively obtaining the satisfied constraints by using a time heuristic search algorithm

Safety trace trj _i And trj _j Comparing the two tracks, and selecting the optimal track to replace the original track of the corresponding vehicle;

s5, repeating the step S4, and determining two vehicles V at the next route point j ═ j +1 _i And V _k If so, planning is repeated until the planning of the traversable tracks of all vehicles is finished or the planning is finished after the maximum planning times is reached, and the planning is taken as an initial solution;

s6, establishing a nonlinear multi-vehicle cooperative control optimization model comprising a target optimization function, vehicle kinematics constraint, barrier collision constraint and inter-vehicle safe distance constraint, taking all vehicle driving tracks as an initial solution, adopting a potential energy risk field method to evaluate vehicle driving risks, and judging two vehicles V _i And V _k Whether or not there is a collision at time T, T0, …, (M-1) × T _d I ≠ 1,2, …, N, k ≠ 1,2, …, N, i ≠ k; if yes, adding the moment into the optimization model as an effective safety distance constraint;

s7, adding all vehicles carrying cargos into the same marshalling to perform track cooperative optimization by using a nonlinear multi-vehicle cooperative control optimization model to obtain the traversable tracks of all vehicles;

the following describes a specific implementation of the above-mentioned unmanned parking path planning method for a special road scene according to a specific embodiment.

Example 1:

the first step is as follows: obtaining starting points P of N vehicles _s,0, ,P _s,1 ,…,P _s,N And its job task end point P _g,0, ,P _g,1 ,…,P _g,N (ii) a Wherein N is a positive integer.

An edge computing unit for a multi-vehicle cooperative control algorithm can be installed in a loading and unloading area, and a global task receiving system in the edge computing unit receives starting points P of N vehicles issued by a cloud intelligent platform _s,0, ,P _s,1 ,…,P _s,N And job task end point P _g,0 ,,P _g,1 ,…,P _g,N And the edge calculation unit records and stores the starting point and the end point of each vehicle task. If the parking point does not exist, only the parking end point is recorded.

The second step is that: and (4) finishing the initial path planning of each vehicle by taking the starting point of each vehicle as a planning initial point and the end point of the operation task as a planning target point by adopting a heuristic search algorithm.

From the first trolley V ₁ Starting planning by using a heuristic search algorithm, taking a starting point as an initial point and an operation task end point as a planning target point, adopting the heuristic search algorithm, and obtaining an initial path pa by avoiding static obstacles in a field in the search process ₁ . Then, for the second trolley V ₂ The path planning is performed in a process similar to that of the first trolley, which is not described herein. And by analogy, finishing the initial path planning of all the vehicles, and thus obtaining the travelable paths of all the vehicles. As shown in FIG. 2, the vehicle V ₁ From P ₀ Starting point along an initial path pa ₁ Travel to job task end point P ₃ Vehicle V ₂ From P ₁ Starting point along an initial path pa ₂ Travel to job task end point P ₆ Vehicle V ₃ From P ₂ Starting point along an initial path pa ₃ Travel to job task end point P ₅ Vehicle V ₄ From P ₃ Starting point along an initial path pa ₄ Travel to job task end point P ₇ Vehicle V ₅ From P ₄ Starting point along an initial path pa ₅ Travel to job task end point P ₀ Vehicle V ₆ From P ₅ Starting point along an initial path pa ₆ Travel to job task end point P ₁ Vehicle V ₇ From P ₆ Starting point along an initial path pa ₇ Travel to job task end point P ₂ Vehicle V ₈ From P ₉ Starting point along an initial path pa ₈ Travel to job task end point P ₄ 。

The third step: acquiring the number M of path points of the longest path in the initial paths of all vehicles, and judging whether the number of the path points of the rest vehicles is less than M or not; if yes, the initial path points are supplemented to M, and the supplemented path coordinate points are the operation task end points; time division is carried out on path points on all the initial paths of the vehicles, and a time sequence T is given _d As adjacent waypoint time intervals.

Specifically, the vehicle V is selected ₁ Initial path pa ₁ At the beginning, the number of recording path points is M. Then, the vehicle V is judged ₂ Initial path pa ₂ Whether the number of waypoints is greater than M. If so, M is pa ₂ If the number of the path points is not the same, the vehicle V is continuously judged ₃ Travelable path pa ₃ Whether the number of waypoints is greater than M. If so, M is pa ₃ The number of waypoints. And by analogy, obtaining the maximum number M of the path points of all the vehicle drivable paths, and completing the path points to M of other vehicle initial paths smaller than the M path points, wherein the completion points correspond to the operation task end points of the vehicle. Then, all the vehicle initial paths are given a time series, T _d Is the time interval between adjacent waypoints. For example, vehicle V ₁ Initial path pa ₁ Time is 0, and the 2 nd is T _d And the 3 rd is 2 x T _d By analogy, the final target point time is (M-1) × T _d 。

The fourth step: judging two vehicles V at the jth path point by taking time as a sequence _i And V _k If j is less than a first threshold value, j is 1, …, M, i is 1,2, …, N, k is 1,2, …, N, i ≠ k; if yes, the driving tracks of the two vehicles are planned again respectively, and safety constraints are added in the planning process

Obtaining satisfied constraints using a time heuristic search algorithm

Safety trajectory trj _i And replace the vehicle V _i The original trajectory; obtaining satisfied constraints using a time heuristic search algorithm

Safety trace trj _j And replace the vehicle V _j The original trajectory;

specifically, it is determined whether there is a collision between the vehicles starting from j equal to 1, when the vehicle V is _i And V _k If collision exists, the heuristic search algorithm is adopted to re-plan the track, and if safety constraint is met

Then the search algorithm selects a parking action to ensure safety during the expansion process. Respectively to vehicle V _i And V _k Planning is carried out, and according to the result, the vehicle track with the shortest track length is selected to replace the original vehicle track.

The fifth step: repeating the fourth step, and judging two vehicles V with the next path point j equal to j +1 _i And V _k If so, repeatedly planning, and if the vehicle V is in the repeated planning process _i Or V _k Planning before the path point j, selecting the path point j as a starting point to plan in the re-planning process, obtaining a planned path by taking an end point as an operation task end point, and repeatedly executing the fourth step until the planning of the traversable tracks of all vehicles is finished or the planning is finished when the maximum planning times is reachedAnd taking it as the initial solution;

and a sixth step: establishing a nonlinear multi-vehicle cooperative control optimization model which comprises a target optimization function, vehicle kinematics constraint, barrier collision constraint and inter-vehicle safe distance constraint, estimating vehicle driving risks by adopting a potential energy risk field method by taking all vehicle drivable tracks as an initial solution, and judging V of two vehicles _i And V _k Whether or not there is a collision at time T, T0, …, (M-1) × T _d I ≠ 1,2, …, N, k ≠ 1,2, …, N, i ≠ k; if yes, adding the moment into the optimization model as an effective safety distance constraint;

specifically, the relative distance and the relative speed between the vehicles are judged, and the driving risk v of the vehicle is evaluated by utilizing a potential energy risk field method _risk If v is _risk >v _thd Wherein v is _thd If the current time is the risk threshold value, taking the time as effective safety cost, and adding the effective safety cost into the optimization model;

the seventh step: adding all vehicles carrying cargos into the same marshalling to perform track cooperative optimization by using a nonlinear multi-vehicle cooperative control optimization model to obtain the traversable tracks of all vehicles;

specifically, all vehicle initial paths are used as model initial solutions, and objective functions related to control quantity input cost and obstacle cost are established; the input cost of the control quantity is the sum of the rotation quantity of a steering wheel, the input quantity of an accelerator pedal and the input quantity of a brake pedal in the process of driving a vehicle along a track; the cost of the obstacle is d _obs And a second threshold value d _thd Difference of (d) _obs Distance between route guide point and nearest obstacle, if d _obs ≥d _thd The obstacle cost is 0, if d _obs ＜d _thd Then the cost of the obstacle is d corresponding to all the route guidance points _obs And d _thd The sum of the squares of the differences; under the condition of meeting the vehicle kinematic model constraint, the control quantity input constraint and the vehicle-to-vehicle safety constraint, the sum of the control quantity input cost and the barrier cost of the target optimization function is minimized, so that a multi-vehicle motion track is obtained; it should be noted that the optimization is preferably performed by selecting the load-carrying vehicle, and when the optimization is completed by the load-carrying vehicle, other empty vehiclesThe obstacle regards the load-carrying vehicle as a detour object, thereby reducing the travel distance of the load-carrying vehicle and improving the economy.

The eighth step: selecting a target with the largest driving risk from the rest vehicles as an object to be optimized, combining the vehicles with the largest collision with the target, and performing track collaborative optimization as a combination to be optimized;

specifically, the driving risk value of the vehicle is obtained according to the seventh step, and the vehicle V with the maximum driving risk is obtained _i Vehicle V most colliding with the vehicle _k As an optimization combination, the two are simultaneously used as a model optimization object to carry out track collaborative optimization; then, V is put _i And V _k The driving track of the vehicle is fixed, and other vehicles need to avoid collision with the two vehicles;

and step nine, repeating the step eight until all the vehicles to be optimized complete the track optimization.

The unmanned multi-vehicle cooperative control method for the loading and unloading scene of the mining area is provided by the invention, and aims to solve the problems of low solving speed and low success rate of unmanned multi-vehicle cooperative control in the loading and unloading scene of the mining area. The multi-vehicle discretization track is obtained based on the conflict resolution and the time-space heuristic search algorithm and is used as an initial solution of the cooperative control model, so that the algorithm solving efficiency can be improved; by establishing a nonlinear multi-vehicle cooperative control optimization model, vehicle grouping and conflict priority is taken as a cooperative control strategy, and the load and unload scene characteristics are combined, so that the convergence speed and the solution success rate of the algorithm can be improved; the driving risk estimation is carried out based on the potential energy risk field, so that the complexity of the cooperative control model can be reduced; by designing a multi-objective optimization function and constraint conditions, the multi-vehicle driving track planning is carried out under the condition of meeting vehicle motion constraint, the vehicle safety distance can be dynamically adjusted according to vehicles with different task functions, the trafficability of group vehicles in a complex and variable environment is improved, and the cooperative loading and unloading operation of a plurality of unmanned vehicles is realized.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (6)

1. An unmanned multi-vehicle cooperative control method for loading and unloading scenes in a mining area is characterized by comprising the following steps:

s1: obtaining starting points P of N vehicles _s,1, ,P _s,2 ,…,P _s,N And its job task end point P _g,1 ,P _g,2 ,…,P _g,N (ii) a Wherein N is a positive integer;

s2: the starting point of each vehicle is taken as a planning initial point, the operation task end point is taken as a planning target point, and the heuristic search algorithm is adopted to complete the planning of the initial path of each vehicle;

s3: acquiring the number M of path points of the longest path in the initial paths of all vehicles, and judging whether the number of the path points of the rest vehicles is less than M or not; if yes, the initial path points are supplemented to M, and the supplemented path coordinate points are the operation task end points; time division is carried out on path points on all the initial paths of the vehicles, and a time sequence T is given _d As adjacent waypoint time intervals;

s4: according to the time sequence, judging two vehicles V at the jth path point _i And V _k If j is less than a first threshold value, j is 1, …, M, i is 1,2, …, N, k is 1,2, …, N, i ≠ k; if yes, the driving tracks of the two vehicles are planned again respectively, and safety constraints are added in the planning process

Respectively obtaining the satisfied constraints by using a time heuristic search algorithm

Safety trace trj _i And trj _j Comparing the two tracks, and selecting the optimal track to replace the original track of the corresponding vehicle;

s5, repeating the step S4, and determining two vehicles V at the next route point j ═ j +1 _i And V _k If so, repeating the planning until the completionPlanning all the vehicle travelable tracks or finishing planning after reaching the maximum planning times, and taking the planned trajectories as an initial solution;

s6, establishing a nonlinear multi-vehicle cooperative control optimization model comprising a target optimization function, vehicle kinematics constraint, barrier collision constraint and inter-vehicle safe distance constraint, taking all vehicle driving tracks as an initial solution, adopting a potential energy risk field method to evaluate vehicle driving risks, and judging two vehicles V _i And V _k At time T, if there is a collision, T is 0, …, (M-1) × T _d I ≠ 1,2, …, N, k ≠ 1,2, …, N, i ≠ k; if yes, adding the moment into the optimization model as an effective safety distance constraint;

and S7, adding all the vehicles with cargos into the same marshalling to perform track cooperative optimization by using a nonlinear multi-vehicle cooperative control optimization model to obtain the driving tracks of all the vehicles.

2. The unmanned multi-vehicle cooperative control method for loading and unloading scenes in mines according to claim 1, wherein in step S2, the initial path planning of each vehicle is completed by using a heuristic search algorithm with the starting point of each vehicle as the planning initial point and the end point of the job task as the planning target point, and specifically comprises:

and generating a drivable path of each vehicle by using a heuristic search algorithm, taking the starting point of each vehicle as a planning initial point, taking the operation task end point as a planning target point, not considering the interactive relation among the vehicles and only considering the obstacles in the scene.

3. The unmanned multi-vehicle cooperative control method for a mine loading and unloading scene according to claim 1, wherein in step S3, the initial path points are supplemented to M, and the supplemented path coordinate points are job task end points; time division is carried out on path points on all the initial paths of the vehicles, and a time sequence T is given _d As the adjacent path point time interval, the method specifically includes:

make up to M initial path points and add T _d As the time interval between adjacent route points, the initial route driving time of all vehicles is (M-1) × T _d 。

4. The unmanned multi-vehicle cooperative control method for loading and unloading scene in mine area as claimed in claim 1, wherein step S4, using time heuristic search algorithm, respectively obtains the satisfaction of constraint

Safety trace trj _i And trj _j Comparing the two tracks, selecting the optimal track to replace the original track of the corresponding vehicle, and specifically comprising the following steps of:

in the time heuristic algorithm searching process, whether the path point j violates the safety constraint condition or not is judged

If yes, the arrival time of the path point is prolonged until the safety constraint condition is met

Respectively planning vehicles V _i And V _k Travel trajectory, comparative trajectory trj _i And trj _k Length of travel time, if trj _i The running time is more than trj _k Then trj _k Replacement vehicle V _k Original trajectory, otherwise trj _i Replacement vehicle V _i The original trajectory.

5. The unmanned multi-vehicle cooperative control method for a mine loading and unloading scene as claimed in claim 1, wherein step S5 is performed to determine the next waypoint j +1 two vehicles V _i And V _k If so, planning repeatedly, and specifically comprising the following steps:

in the repeated planning process, if the vehicle V _i Or V _k Planning is carried out before the path point j, the path point j is selected as a starting point to carry out planning in the next planning process, and the end point is an operation task end point.

6. The unmanned parking path planning method for special road scenes as claimed in any one of claims 1 to 5, wherein in step S7, all vehicles with cargos are added into the same group by using a nonlinear multi-vehicle cooperative control optimization model for track cooperative optimization to obtain the traversable tracks of each vehicle, further comprising:

and selecting the target with the maximum driving risk from the rest vehicles as an object to be optimized, combining the vehicles with the maximum collision with the target, and performing track collaborative optimization as a combination to be optimized.

Images