Robot path planning method based on IBA*-DWA fusion algorithm
Through the improved fusion algorithm of two-way A* and dynamic window method (IBA*-DWA), the problems of low efficiency and high energy consumption of traditional algorithms in complex environments are solved, and more efficient and better path planning and Energy consumption is reduced.
Patent Information
- Application Number
- CN202410476820.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-19
- Publication Date
- 2024-11-15
AI Technical Summary
When traditional bidirectional A* algorithm encounters many obstacles on the connection between the starting node and the target node, the path planning efficiency is low and dynamic obstacles cannot be avoided; traditional DWA algorithms are prone to falling into local optimality and cannot reach the target point. And high energy consumption.
Using the improved fusion algorithm of two-way A* and dynamic window method (IBA*-DWA), the two-way A* algorithm is improved through the antagonistic attraction strategy, the environmental condition coefficient and speed driving strategy are increased, and the path planning is optimized to ensure that the robot can effectively avoid it. Dynamic obstacles and reduce energy consumption.
Improve the efficiency and quality of path planning, ensure that the robot can effectively avoid dynamic obstacles, reduce energy consumption, and ensure that it can reach the target point, providing a better global path.
Smart Images

Figure CN118963333A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of robot path planning technology, and is particularly a method based on an improved dynamic window approach (IDWA) and an improved bidirectional A * (improved Bidirectional A * , IBi-A * ) fusion algorithm (IBA * -DWA) is used for robot path planning. Background Art
[0002] Path planning is the core technology of mobile robot navigation. It means that the mobile robot plans an optimal path from the initial point to the target point based on the evaluation index from the current environment. However, in the actual complex and changeable environment, there will be interference from moving obstacles such as vehicles and pedestrians on the mobile robot's forward path, which can easily lead to collisions. Therefore, it is necessary to propose an algorithm that takes into account both the global optimal planning path and the local dynamic obstacle avoidance path.
[0003] At present, the most widely used global path planning algorithms are Dijkstra algorithm, simulated annealing method, genetic algorithm, A * Algorithm, the commonly used local path planning algorithms include artificial potential field method, dynamic window method, time elastic band method and so on.
[0004] Among them A * The algorithm is a classic heuristic search algorithm that can quickly find the shortest path and is widely used in global path planning. * The algorithm also has the problem of traversing too many nodes, and the bidirectional A using the bidirectional search method * The algorithm can effectively reduce the number of nodes traversed, but the traditional two-way A * When the algorithm encounters a situation where there are many obstacles on the line connecting the starting node and the target node, the search directions of the forward search and the backward search may be different, resulting in the forward search and the backward search not being able to meet normally in the path, or even not meeting until near the end point, reducing the path-finding efficiency and path quality. *The algorithm is only applicable to static environments and cannot avoid dynamic obstacles. Therefore, this application integrates the dynamic window method to perform local path planning based on the global path. The dynamic window method is a classic local path planning algorithm. The dynamic window method has the advantages of simple model, relatively smooth path, strong practicality, real-time obstacle avoidance function, and considering the performance of the robot itself. Therefore, it is widely used in local path planning. However, the traditional DWA algorithm is prone to fall into the local optimum near the target point due to the influence of obstacles, and may even fail to reach the target point. In addition, when the traditional DWA algorithm faces certain dynamic obstacles, the angle between the obstacle's movement direction and the robot's detour direction to avoid the obstacle is too small and the speeds of the two are similar, resulting in the robot's detour path being too long or even driving beyond the map range. Summary of the invention
[0005] The object of the present invention is to provide a method based on an improved dynamic window approach (IDWA) and an improved bidirectional A * (improved Bidirectional A * , IBi-A * ) fusion algorithm (IBA * -DWA) is used for robot path planning. This method can overcome the traditional two-way A * The algorithm has a problem of reduced path-finding efficiency when traversing many nodes and solving the path planning problem of many obstacles in the connecting area between the starting node and the target node. At the same time, this method not only overcomes the problem that the traditional DWA cannot reach the target node, is prone to fall into the local optimum, and the driving path is overly affected by dynamic obstacles, but also reduces the energy consumption of the robot. Finally, the two algorithms are integrated, which not only improves the global path search efficiency, but also realizes real-time obstacle avoidance based on the optimal path.
[0006] To solve the above problems, the present invention proposes a method based on an improved dynamic window approach (IDWA) and an improved bidirectional A * (improved Bidirectional A * , IBi-A * ) fusion algorithm (IBA * -DWA) is used for robot path planning.
[0007] IBA * -DWA first proposed a different-direction attraction strategy for bidirectional A when performing global path planning. * The improved algorithm can effectively avoid the problem of different directions of forward search and backward search and improve search efficiency.
[0008] When using DWA for local path planning, it is proposed to increase the environmental condition coefficient so that when near the target node, the weights of the sub-functions in the evaluation function can be adjusted autonomously according to the actual environmental conditions to avoid falling into the local optimum or even being unable to reach the target point. At the same time, a speed-changing driving strategy is proposed, which can effectively avoid dynamic obstacles and drive along the global optimal path to the maximum extent while reducing the robot's energy consumption.
[0009] A large number of simulation results show that the IBA adopting the new strategy of this application * -The DWA algorithm has made significant progress in solving robot path planning problems, and its overall performance is better than traditional algorithms.
[0010] Two-way A * The algorithm starts searching from both the starting point and the end point, and uses heuristic functions to evaluate the priority of the nodes. In each step, the node with the lowest priority is selected from both directions for expansion. The search from the starting point to the end point is called forward search, and the search from the end point to the starting point is called backward search. The improved bidirectional A* algorithm of the present invention proposes to change the calculation of the distance from the current point to the end point in the forward search heuristic function to the calculation of the distance from the current point to the current node of the backward search. Similarly, the calculation of the distance from the current point to the starting point in the backward search heuristic function is changed to the calculation of the distance from the current point to the current node of the forward search, so that the forward search and the backward search are dynamically attracted to each other. The current point of the forward search is (x0, y0), and the current point of the backward search is (x1, y1). The improved heuristic function is as formula (1), and the estimated path evaluation function is as formula (2).
[0011]
[0012]
[0013] In formula (1), g(n0) and g(n1) are the forward search true path evaluation function and the backward search true path evaluation function respectively.
[0014] In order to verify the feasibility of the anisotropic attraction strategy proposed in this paper, Python language programming was used to test the bidirectional A with only the improved anisotropic attraction strategy in a 20*20 environment model with static obstacles. * algorithm is simulated and compared with the basic bidirectional A * The simulation results are shown in Figure 2. Figure 2 , Figure 3 As shown, the star-shaped nodes are forward search traversal nodes, and the circular nodes are backward search access nodes. It can be seen that the basic bidirectional A *When the algorithm is performing path planning, the forward search and backward search are in different directions, resulting in the backward search not finding the final path until near the starting point. In the improved algorithm, the forward search is attracted by the backward search, the path finding direction is consistent, the path quality is better, and the path finding efficiency is higher.
[0015] Table 1 Comparison of algorithm experimental data under M1 environment
[0016]
[0017] The present invention uses a known redundant point deletion strategy to process the global path obtained by the improved bidirectional A* algorithm, deletes the redundant turning points in the path, and leaves the key nodes as the sub-target nodes of the local path planning of the improved DWA algorithm, that is, the path is segmented according to the key nodes, and the local path planning is performed in sequence until the final target node. The specific idea of the redundant node deletion strategy is to connect the first point and the second point of the global path. If there is no obstacle between the two, then delete the second point; then connect the first point and the third point. If there is no obstacle between the two, then delete the third point; and so on, until an obstacle appears between the first point and the point connected to it, retain the point as a key node, and then continue the above operations from this point until the end point, and the nodes left at the end are all the key nodes.
[0018] Traditional DWA determines a sampling speed space that meets the hardware constraints of the mobile robot in the speed space according to the current position state and speed state of the mobile robot, performs sampling in the sampling speed space, generates a series of candidate speeds, and then calculates the trajectory of the mobile robot moving within a certain period of time under these speed conditions, and evaluates the trajectory through an evaluation function, and finally selects the speed corresponding to the best evaluated trajectory as the movement speed of the mobile robot.
[0019] In view of the problems that the traditional DWA algorithm cannot reach the end point and the robot consumes a lot of energy, it is proposed to add the environmental condition coefficient to the trajectory evaluation function, and take the robot as the core and D as the radius as the calculation range. The evaluation function of the improved DWA algorithm is as follows:
[0020]
[0021] Where H(v,ω) is the heading angle sub-function, S(v,ω) is the speed sub-function, D(v,ω) is the obstacle distance sub-function, α, β, γ are the weight values of the corresponding sub-functions, l1 and l2 are environmental condition coefficients, defined as in formula (4):
[0022]
[0023] Where δ is a positive number less than 1, D is the diameter of the calculation range, d1 is the distance from the robot to the obstacle, and d2 is the distance from the robot to the target node.
[0024] When both the obstacle and the target node are out of the calculation range, the heading angle is preferred as the guidance; if the target node is out of the calculation range but the obstacle is in the calculation range, the obstacle distance is preferred as the guidance; if the target node is in the calculation range, the heading angle is preferred as the guidance.
[0025] By conducting comparative simulation experiments under a 10*10 static environment model, the experimental results are as follows Figure 4 , Figure 5 It can be seen that the traditional DWA cannot reach the end point due to the influence of obstacles around the end point, while the improved DWA can combine the actual environment and increase the influence weight of the end point so that the robot can reach the end point.
[0026] In order to solve the problem of the robot taking detours to avoid dynamic obstacles while reducing the robot's energy consumption, the present invention proposes that when it is recognized that a mobile obstacle will pass in the robot's driving direction and may collide with the mobile robot, the distance between the current position and the possible collision point is calculated, and the driving speed is reduced in advance to ensure that the robot will not encounter obstacles when driving at this speed, so that the robot can pass the obstacle at a low speed and drive according to the global path to the maximum extent, reducing detours caused by obstacle avoidance, and at the same time reducing the number of times the robot brakes and starts to reduce energy consumption.
[0027] In order to verify the effectiveness of the speed change strategy, a simulation test was conducted in a 10*10 environment model to compare it with the traditional DWA. Figure 6 , Figure 8 are the operating states of the two algorithms at the same four time points. Figure 7 , Fig. 9 The speed and angular velocity of the two algorithms change with time (the upper curve in the figure is the speed change curve, and the lower curve is the angular velocity change curve). Figure 6 , Figure 7 The traditional DWA algorithm chooses to detour when encountering dynamic obstacles. Because the obstacles are dense, the robot's detour route is blocked, resulting in a very long detour route, a large angular velocity change, and a longer running time. Figure 8 , Fig. 9 The improved DWA algorithm in the algorithm chooses to slow down and keep moving slowly in sync with the obstacles. After avoiding the obstacles, it finally reaches the destination along the original route. The final path is shorter and smoother.
[0028] The technical solution adopted by the present invention to solve its technical problem is:
[0029] An IBA-based *-Robot path planning based on DWA fusion algorithm includes the following steps:
[0030] S1. Initialize algorithm-related parameters, including: map data map, starting point s, end point e, heuristic function type heuristic_type, initialize OPEN list, CLOSED list and parent node list for forward search and backward search respectively, path list path, dynamic obstacle speed dynamic_obstacle_speed, and the initial state of the robot, including direction angle (θ), speed (v), and rotation speed (ω);
[0031] S2, for both directions A * The algorithm is improved. According to the improved heuristic function formula (1), forward and backward searches are started simultaneously until the nodes in the two search queues meet or one of the queues is empty. The forward search starts from the starting point and searches toward the end point. Each time, the node with the smallest f value in the OPEN list is selected for expansion. For the expanded node, the cost of its adjacent nodes is calculated, and the OPEN and CLOSED lists are updated. The backward search starts from the end point and searches toward the starting point. Similarly, the node with the smallest f value is selected for expansion, and the OPEN and CLOSED lists are updated.
[0032]
[0033] Where g(n0) and g(n1) are the forward search true path evaluation function and the backward search true path evaluation function respectively, and the estimated path evaluation function h(n) is as shown in formula (2):
[0034]
[0035] The current point of the forward search is (x0, y0), and the current point of the backward search is (x1, y1);
[0036] S3. During the forward and backward search process, check whether the nodes meet. If a node exists in both the CLOSED list of the forward search and the parent node list of the backward search, it means that the two search queues meet.
[0037] S4, during the forward and backward search process, check if the nodes meet, execute S5, if no nodes meet, execute S2;
[0038] S5. Starting from the meeting point, trace back to the starting point and the end point respectively through the parent node dictionary of the forward search queue and the parent node list of the backward search queue to extract the complete path;
[0039] S6, from the improved two-way A * A series of key nodes are extracted from the complete path obtained by the algorithm;
[0040] S7, pass in the key node as a parameter and execute the improved DWA algorithm;
[0041] S7-1, based on the existing DWA algorithm, including: calculating the range limits of the speed and rotation speed of the robot in the next sampling cycle, sampling in the speed space of the robot, obtaining path information and evaluation factors, and generating a set of candidate speeds;
[0042] S7-2, the improvement part includes, for each candidate speed, calculating the braking distance of the robot and scoring it, evaluating the quality of the generated trajectory according to the evaluation function formula (3) based on the robot's dynamic model and speed change obstacle avoidance strategy, and selecting the best trajectory from the evaluated trajectories as the action to be performed by the robot at the next moment;
[0043] Eval(v,ω)=l1αH(v,ω)+βS(v,ω)+l2γD(v,ω) (3)
[0044] Where H(v,ω) is the heading angle sub-function, S(v,ω) is the speed sub-function, D(v,ω) is the obstacle distance sub-function, α, β, γ are the weight values of the corresponding sub-functions, l1 and l2 are environmental condition coefficients, where l1 and l2 are environmental condition coefficients, defined as formula (4):
[0045]
[0046] Where δ is a positive number less than 1, D is the diameter of the calculation range, d1 is the distance from the robot to the obstacle, and d2 is the distance from the robot to the target node;
[0047] S8, simulate the movement of the robot in the environment, update the robot status, and dynamically update the position of obstacles during the movement of the robot;
[0048] S9, record the robot's motion trajectory and determine whether it has reached the end point. If it has, stop the simulation and output the final path. If it has not reached the end point, execute S7-1.
[0049] The beneficial effect of the present invention is to provide a method based on IBA * -DWA fusion algorithm is used for robot path planning. This method can overcome the traditional two-way A * The algorithm cannot avoid dynamic obstacles, traverses too many nodes, and has poor path quality. At the same time, it solves the problems of DWA being unable to reach the end point, taking detours to avoid dynamic obstacles, and high robot energy consumption. It improves the efficiency of global path planning and the quality of the final path, and reduces the energy consumption of the robot.
[0050] Improved Two-Way A *The algorithm adds the search situation in the other direction to the evaluation function calculation of this direction, allowing the forward and backward directions to attract each other, improving the search efficiency and the global path quality. It adds the environmental condition coefficient to the DWA trajectory evaluation function, solving the problem that the robot is trapped in the local optimum and cannot reach the end point. The speed change strategy enables the robot to travel along the global optimal path to the maximum extent, reducing the number of brake starts caused by obstacle avoidance in the middle, improving the path quality, and reducing the robot's energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] The present invention is further described below in conjunction with the accompanying drawings and embodiments:
[0052] Figure 1 It is a kind of IBA based on the present invention * - Algorithm flow chart of robot path planning based on DWA algorithm;
[0053] Figure 2 Traditional two-way A * Algorithm optimal path diagram;
[0054] Figure 3 Improved Two-Way A * Algorithm optimal path diagram;
[0055] Figure 4 Final path diagram of the traditional DWA algorithm in a static environment;
[0056] Figure 5 Improved final path map of DWA algorithm in static environment;
[0057] Figure 6 Four moment path diagrams of the traditional DWA algorithm in a dynamic environment;
[0058] Figure 7 The curve of velocity and angular velocity changing with time of the traditional DWA algorithm in dynamic environment;
[0059] Figure 8 Improved DWA algorithm 4-moment path diagram in dynamic environment;
[0060] Fig. 9 Improved DWA algorithm velocity and angular velocity curves over time in dynamic environment;
[0061] Fig.10 Final path diagram of traditional DWA in 20×20 environment;
[0062] Fig.11 Final path diagram of the fusion algorithm in a 20×20 environment;
[0063] Fig.12 The final path diagram of the improved algorithm in the literature and the algorithm of the present invention. DETAILED DESCRIPTION
[0064] See also Figure 1 The present invention provides a method based on IBA * -Robot path planning based on DWA fusion algorithm includes the following steps:
[0065] S1. Initialize algorithm-related parameters, including: map data map, starting point s, end point e, heuristic function type heuristic_type, initialize OPEN list, CLOSED list and parent node list for forward search and backward search respectively, path list path, dynamic obstacle speed dynamic_obstacle_speed, and the initial state of the robot, including direction angle (θ), speed (v), and rotation speed (ω);
[0066] S2, for both directions A * The algorithm is improved. According to the improved heuristic function formula (1), forward and backward searches are started simultaneously until the nodes in the two search queues meet or one of the queues is empty. The forward search starts from the starting point and searches toward the end point. Each time, the node with the smallest f value in the OPEN list is selected for expansion. For the expanded node, the cost of its adjacent nodes is calculated, and the OPEN and CLOSED lists are updated. The backward search starts from the end point and searches toward the starting point. Similarly, the node with the smallest f value is selected for expansion, and the OPEN and CLOSED lists are updated.
[0067]
[0068] Where g(n0) and g(n1) are the forward search true path evaluation function and the backward search true path evaluation function respectively, and the estimated path evaluation function h(n) is as shown in formula (2):
[0069]
[0070] The current point of the forward search is (x0, y0), and the current point of the backward search is (x1, y1);
[0071] S3. During the forward and backward search process, check whether the nodes meet. If a node exists in both the CLOSED list of the forward search and the parent node list of the backward search, it means that the two search queues meet.
[0072] S4, during the forward and backward search process, check if the nodes meet, execute S5, if no nodes meet, execute S2;
[0073] S5. Starting from the meeting point, trace back to the starting point and the end point respectively through the parent node dictionary of the forward search queue and the parent node list of the backward search queue to extract the complete path;
[0074] S6, from the improved two-way A * A series of key nodes are extracted from the complete path obtained by the algorithm;
[0075] S7, pass in the key node as a parameter and execute the improved DWA algorithm;
[0076] S7-1, based on the existing DWA algorithm, including: calculating the range limits of the speed and rotation speed of the robot in the next sampling cycle, sampling in the speed space of the robot, obtaining path information and evaluation factors, and generating a set of candidate speeds;
[0077] S7-2, the improvement part includes, for each candidate speed, calculating the braking distance of the robot and scoring it, evaluating the quality of the generated trajectory according to the evaluation function formula (3) based on the robot's dynamic model and speed change obstacle avoidance strategy, and selecting the best trajectory from the evaluated trajectories as the action to be performed by the robot at the next moment;
[0078] Eval(v,ω)=l1αH(v,ω)+βS(v,ω)+l2γD(v,ω) (3)
[0079] Where H(v,ω) is the heading angle sub-function, S(v,ω) is the speed sub-function, D(v,ω) is the obstacle distance sub-function, α, β, γ are the weight values of the corresponding sub-functions, l1 and l2 are environmental condition coefficients, where l1 and l2 are environmental condition coefficients, defined as formula (4):
[0080]
[0081] Where δ is a positive number less than 1, D is the diameter of the calculation range, d1 is the distance from the robot to the obstacle, and d2 is the distance from the robot to the target node;
[0082] S8, simulate the movement of the robot in the environment, update the robot status, and dynamically update the position of obstacles during the movement of the robot;
[0083] S9, record the robot's motion trajectory and determine whether it has reached the end point. If it has, stop the simulation and output the final path. If it has not reached the end point, execute S7-1.
[0084] The effect of the present invention can be further illustrated by the following simulation experiment:
[0085] In order to verify the correctness and rationality of this method, Python language programming is used to simulate the environment model M1 with 20×20 dynamic obstacles and static obstacles, and compared with the traditional DWA algorithm. The simulation results are shown in the figure. Fig.10 , Fig.11 shown. Fig.10 , Fig.11 The middle circle is a static obstacle, the square is a dynamic obstacle, and the dynamic obstacle moves from the left side of the map to the right. The lower left corner is the starting point, and the five-pointed star in the upper right corner is the end point. The triangle represents the robot, the solid line represents the path the robot has walked, and the dotted line represents the two-way A * The global optimal path obtained by the algorithm is Fig.10 It can be seen that the final path is very tortuous, and a large part of redundant paths appear in the middle due to avoiding obstacles. Fig.11 The robot not only drives along the global optimal path, but also optimizes the path based on the global optimal path, making the path shorter and smoother.
[0086] In order to further verify the stability of the improved algorithm proposed in the present invention, the hybrid algorithm of improved particle swarm differential evolution and dynamic window method recorded in the article "Dynamic Path Planning of Improved Particle Swarm Optimization and Dynamic Window Method" published in the journal "Small and Microcomputer Systems" in 2023 was selected. In the 10×10 grid environment of the article, the method of the present invention was used for simulation.
[0087] The simulation results are as follows Fig.12 As shown, the circles are static obstacles, the squares are dynamic obstacles, and the dynamic obstacles move from the left side of the map to the right, with the lower left corner as the starting point. Fig.12 (a) is the simulation result in the original document, the final path length of which is 18.7m. Fig.12 (b) is the final path obtained by the algorithm of the present invention. The length of the final path of the present invention is 18.68m. It can be seen from the figure that the final path obtained by the improved algorithm in the literature has too many turning points, and the path is close to obstacles in many places. The final path obtained by the present invention has a larger corner angle, a smoother path, no extra turning points, and is closer to the actual path.
Claims
1. An IBA-based * -Robot path planning based on DWA fusion algorithm includes the following steps: S1. Initialize algorithm-related parameters, including: map data map, starting point s, end point e, heuristic function type heuristic_type, initialize OPEN list, CLOSED list and parent node list for forward search and backward search respectively, path list path, dynamic obstacle speed dynamic_obstacle_speed, and the initial state of the robot, including direction angle (θ), speed (v), and rotation speed (ω); S2, for both directions A * The algorithm is improved. According to the improved heuristic function formula (1), forward and backward searches are started simultaneously until the nodes in the two search queues meet or one of the queues is empty. The forward search starts from the starting point and searches toward the end point. Each time, the node with the smallest f value in the OPEN list is selected for expansion. For the expanded node, the cost of its adjacent nodes is calculated, and the OPEN and CLOSED lists are updated. The backward search starts from the end point and searches toward the starting point. Similarly, the node with the smallest f value is selected for expansion, and the OPEN and CLOSED lists are updated. Where g(n0) and g(n1) are the forward search true path evaluation function and the backward search true path evaluation function respectively, and the estimated path evaluation function h(n) is as shown in formula (2): Where the current point for forward search is (x0, y0), and the current point for backward search is (x1, y1); S3. During the forward and backward search process, check whether the nodes meet. If a node exists in both the CLOSED list of the forward search and the parent node list of the backward search, it means that the two search queues meet. S4, during the forward and backward search process, check if the nodes meet, execute S5, if no nodes meet, execute S2; S5. Starting from the meeting point, trace back to the starting point and the end point respectively through the parent node dictionary of the forward search queue and the parent node list of the backward search queue to extract the complete path; S6, from the improved two-way A * A series of key nodes are extracted from the complete path obtained by the algorithm; S7, pass in the key node as a parameter and execute the improved DWA algorithm; S7-1, based on the existing DWA algorithm, including: calculating the range limits of the speed and rotation speed of the robot in the next sampling cycle, sampling in the speed space of the robot, obtaining path information and evaluation factors, and generating a set of candidate speeds; S7-2, the DWA algorithm improvement part includes, for each candidate speed, calculating the braking distance of the robot and scoring it, evaluating the quality of the generated trajectory according to the evaluation function formula (3) based on the robot's dynamic model and speed change obstacle avoidance strategy, and selecting the best trajectory from the evaluated trajectories as the action to be performed by the robot at the next moment; Eval(v,ω)=l1αH(v,ω)+βS(v,ω)+l2γD(v,ω) (3) Where H(v,ω) is the heading angle sub-function, S(v,ω) is the speed sub-function, D(v,ω) is the obstacle distance sub-function, α, β, γ are the weight values of the corresponding sub-functions, l1 and l2 are environmental condition coefficients, where l1 and l2 are environmental condition coefficients, defined as formula (4): Where δ is a positive number less than 1, D is the diameter of the calculation range, d1 is the distance from the robot to the obstacle, and d2 is the distance from the robot to the target node; S8, simulate the movement of the robot in the environment, update the robot status, and dynamically update the position of obstacles during the movement of the robot; S9, record the robot's motion trajectory and determine whether it has reached the end point. If it has, stop the simulation and output the final path. If it has not reached the end point, execute S7-1.
Citation Information
Cited By
Robot path planning method and system, medium, product and computer equipment
CN120029302A
