Distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA

Abstract

The invention discloses a distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA, and belongs to the technical field of path planning. The invention discloses a distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA. The method comprises the following steps: constructing an environment model; performing global path planning by using an ant colony algorithm ACO; establishing a motion model of the robot, and selecting an optimal speed group; and designing a conflict judgment rule, introducing a priority strategy, and solving conflicts according to the importance and urgency of robot tasks. According to the distributed multi-mobile-robot path planning and obstacle avoidance method based on ACO-DWA, an improved dynamic window method (DWA) and an ant colony optimization (ACO) algorithm are utilized, effective local path planning and global path optimization are achieved in the environment with high-density obstacles and unknown static obstacles, and safe obstacle avoidance of the robots is ensured. Meanwhile, a motion conflict judgment rule and a priority strategy in the multi-robot system are provided to coordinate the motion between the robots and between the robots and obstacles, and safe and efficient navigation is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of path planning, and in particular to a distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA. Background Art

[0002] Mobile robots, with their autonomous navigation and flexible movement, have been increasingly widely used in many fields such as logistics, daily services, and industrial manufacturing. With the development of technology, facing more stringent requirements and more complex environments, a single mobile robot has been unable to complete many tasks, and multiple mobile robots have entered the field of vision of researchers. Compared with a single robot, a multi-mobile robot system (MRSs) has the advantages of good flexibility and a wide range of exploration, and can complete complex tasks faster and better than a single robot. Research on distributed multi-robot path planning and obstacle avoidance methods can effectively improve the real-time obstacle avoidance ability of robots in complex environments, which helps to improve production efficiency and reduce operational risks.

[0003] Traditional multi-robot navigation methods are mainly related to reinforcement learning, neural networks and deep learning, which are all affected by demanding data sets and training models. More importantly, the robot dynamics constraints are ignored. In view of this, the present invention proposes a distributed multi-robot path planning and obstacle avoidance method based on an ant colony algorithm fused with a dynamic window method, which does not require excessive reliance on prior information for decision-making, and can also achieve optimal path planning and real-time obstacle avoidance of robots in complex environments.

[0004] The ACO algorithm is suitable for global path planning, can find the optimal path in a complex environment, adapt to dynamic changes, and can be extended to a multi-robot system to deal with path conflicts. The DWA algorithm focuses on local path planning, emphasizes real-time and obstacle avoidance capabilities, can quickly calculate safe motion trajectories, and is suitable for navigation in dynamic environments. The combination of the two can achieve efficient and safe robot path planning and obstacle avoidance, and enhance the robot's navigation capabilities. The present invention also combines the robot kinematic model for analysis to provide a new method for distributed multi-mobile robot path planning and obstacle avoidance. Summary of the invention

[0005] The purpose of this invention is to provide a distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA, which uses the improved dynamic window method (DWA) and ant colony optimization (ACO) algorithm to achieve effective local path planning and global path optimization in an environment with high-density obstacles and unknown static obstacles. At the same time, motion conflict judgment rules and priority strategies in multi-robot systems are proposed to coordinate the motion between robots and obstacles to ensure safe and efficient navigation.

[0006] To achieve the above object, the present invention provides a distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA, comprising the following steps:

[0007] S1: Use sensors to obtain global environmental information and build an environmental model; the global environmental information includes static obstacle distribution and dynamic obstacle distribution;

[0008] S2: Use the ant colony algorithm ACO for global path planning to optimize path length and number of turns;

[0009] S3: Establish a robot motion model. In local path planning, based on the robot motion model, sample multiple speed combinations, analyze and predict the trajectory of each speed combination, evaluate the quality of the trajectory, and select the best speed group.

[0010] S4: In a multi-robot system, design conflict judgment rules to identify potential motion conflicts between robots; introduce a priority strategy to resolve conflicts based on the importance and urgency of the robot tasks.

[0011] Preferably, the S1 specifically comprises the following steps:

[0012] S11, using a grid method to divide the entire free space into grid cells of uniform size to form a matrix MAP;

[0013] S12, marking the matrix MAP with environmental information according to the acquired global environmental information;

[0014] S13, marking a serial number for each grid cell in the matrix MAP;

[0015] S14. The robot obtains environmental information through sensors and dynamically updates the environmental information of each grid unit.

[0016] Preferably, the S2 specifically comprises the following steps:

[0017] S21. Improve the initial pheromone distribution and allocate pheromone concentration values ​​on the grid cells according to the percentage of obstacles around the grid cell nodes.

[0018] τ ij (0) = ε*f(j)

[0019]

[0020] Among them, τ ij is the pheromone concentration value on the path from node i to node j, ε is the enhancement factor of the initial pheromone, f(j) represents the obstacle avoidance factor, C U is the complementary set symbol, allowed jRepresents the number of obstacle grids around point j; the fewer the number of obstacle grids in the surrounding grid cells, the more significant the initial pheromone of the grid;

[0021] S22. The probability of the robot selecting the next node is:

[0022]

[0023] in, is the probability that robot k moves from node i to node j at time t; η ij (t) is the heuristic function; the robot randomly selects a path based on the selection probability and performs global path planning;

[0024] S23. After all robots complete path selection, they update the pheromones on the path. The pheromone update strategy is:

[0025] τ ij (t+1)=(1-ρ)τ ij (t)+Δτ ij (t,t+1)

[0026]

[0027] Among them, ρ is the pheromone volatility coefficient, Δτ ij is the pheromone increment during the robot's path-finding process, Q is the pheromone intensity, and L is the total path length of robot k;

[0028] S24. Compare the path qualities of all robots, obtain the optimal path and its corresponding path length within the iteration number, and use the optimal path and path length within the iteration number as a reference for the next round of iteration to optimize the path; after the stopping condition is met, output the optimal path and the length corresponding to the optimal path as the final global path planning result.

[0029] Preferably, in S3, the robot motion model is

[0030]

[0031] Where v(t) is the linear velocity of the robot, ω(t) is the angular velocity of the robot, x(t) and y(t) represent the positions of the robot on the x-axis and y-axis at time t, respectively, θ(t) is the orientation of the robot, and Δt is the time step.

[0032] Preferably, in S3, the speed constraint of the robot is:

[0033] v s ={(v,ω)|v∈[v min ,v max ],ω∈[ω min ,ωmax ]}.

[0034] Preferably, in S3, the acceleration constraint of the robot is:

[0035]

[0036] in, and are the minimum linear deceleration and minimum angular deceleration respectively; and are the maximum linear acceleration and maximum angular acceleration respectively.

[0037] Preferably, in S3, the braking distance constraint of the robot is:

[0038]

[0039] Where dist(v,ω) is the distance between the simulated trajectory and the nearest obstacle.

[0040] Preferably, in S3, the speed combination evaluation function is:

[0041] G(v,ω)=σ(A·heading(v,ω)+B·dist(v,ω)+C·velocity(v,ω))

[0042] Among them, G(v,ω) is the evaluation value of the simulated trajectory, heading(v,ω) represents the angle between the robot’s heading and the target line, dist(v,ω) is the distance between the robot and the nearest obstacle, velocity(v,ω) is used to evaluate the robot’s motion performance, A, B, and C are the weight factors of the corresponding evaluation functions, and σ is the normalization factor.

[0043] Preferably, in S4, the conflict judgment rule design process specifically includes the following steps:

[0044] S41. Set a corresponding priority for each robot;

[0045] S42. When the actual distance between the two robots is less than the expected obstacle avoidance distance, the low-priority robot establishes a local coordinate axis, calculates the angle α between the straight line of the two robots and the positive direction of the x-axis, and then calculates the difference between α and the orientation angle β of the high-priority robot.

[0046] S43, when |α-β|<90°, the low-priority robot decelerates and stops in a short time, and the high-priority robot regards the low-priority robot as an unknown static obstacle and performs obstacle avoidance movement through DWA.

[0047] The advantages and positive effects of the distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA described in the present invention are:

[0048] 1. The present invention adopts the ACO-DWA fusion algorithm and combines the priority strategy for multi-robot path planning and obstacle avoidance, which can effectively combine the advantages of global optimization and local real-time response to ensure the safety of robot obstacle avoidance. The priority strategy can effectively schedule multiple robots, reduce path conflicts, improve resource utilization and overall system efficiency. It not only improves the flexibility and reliability of the robot, but also reduces energy consumption, optimizes the overall operation effect, and adapts to the task requirements in complex dynamic environments.

[0049] 2. The ACO-DWA fusion algorithm and priority strategy are used for multi-robot path planning and obstacle avoidance, which significantly improves the operation efficiency and resource utilization, reduces the risk of collision, and opens up new ideas for the research on the collaboration and dynamic environment adaptation of multi-agent systems, which has important theoretical and practical significance.

[0050] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a flow chart of an embodiment of the present invention;

[0052] Figure 2 This is a flow chart of optimal global path planning using the ant colony algorithm according to an embodiment of the present invention;

[0053] Figure 3 This is a schematic diagram of the initial distribution of pheromones according to an embodiment of the present invention;

[0054] Figure 4 A flowchart of local optimal path planning provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.

[0056] like Figure 1 A distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA includes the following steps:

[0057] S1: Use sensors to obtain global environmental information and build an environmental model; the global environmental information includes static obstacle distribution and dynamic obstacle distribution.

[0058] S1 specifically includes the following steps:

[0059] S11. Use a grid method to divide the entire free space into grid cells of uniform size to form a matrix MAP.

[0060] S12. Marking the environment information of the matrix MAP according to the acquired global environment information.

[0061] S13. Mark each grid cell in the matrix MAP with a serial number to facilitate subsequent acquisition of the actual coordinates of the grid cell.

[0062] S14. During the actual operation, the robot obtains environmental information through sensors and dynamically updates the environmental information of each grid unit to adapt to environmental changes.

[0063] S2: Use the ant colony algorithm ACO for global path planning and optimize the path length and number of turns.

[0064] Figure 2 The following is a flowchart of the optimal global path planning using the ant colony algorithm according to an embodiment of the present invention. Figure 2 As shown in the figure, the use of ant colony algorithm to achieve global optimal path planning mainly includes the following steps: first, initialize the robot's position and set pheromone; then, the robot selects a path based on the pheromone concentration and heuristic function; then, the robot traverses all nodes and records the path; finally, update the pheromone to reflect the quality of the path, repeat the iteration until the termination condition is met, and finally obtain the optimal path.

[0065] S2 specifically includes the following steps:

[0066] S21. Set parameters, including information heuristic factor α, expected heuristic factor β, pheromone volatility coefficient ρ, etc. In the path network, each edge (connection between nodes) initializes the pheromone concentration value to τ. Set the total number of iterations of the algorithm.

[0067] Improve the initial pheromone distribution and assign pheromone concentration values ​​on the grid cells according to the percentage of obstacles around the grid cell nodes.

[0068] τ ij (0) = ε*f(j)

[0069]

[0070] Among them, τ ij is the pheromone concentration value on the path from node i to node j, ε is the enhancement factor of the initial pheromone, f(j) represents the obstacle avoidance factor, C U is the complementary set symbol, allowed jRepresents the number of obstacle grids around point j. Using the function f(j), the percentage of free grids in the 8 adjacent grids of the current grid can be calculated. When the number of obstacle grids in the adjacent grids is smaller, the initial pheromone of the grid is more significant, and vice versa, the initial pheromone is smaller. This processing of pheromone distribution can guide the robot to move forward, avoid the robot from performing too much terrain pathfinding for obstacles, and speed up the convergence of the algorithm.

[0071] Figure 3 This is a schematic diagram of the initial distribution of pheromones according to an embodiment of the present invention.

[0072] S22, robot movement. Each robot starts from the starting node S, and the probability of the robot selecting the next node is:

[0073]

[0074] in, is the probability that robot k moves from node i to node j at time t; η ij (t) is a heuristic function, which is usually inversely proportional to the distance between nodes.

[0075] The robot randomly selects a path based on the selection probability and performs global path planning. Each robot avoids nodes that have been visited when selecting a path until it reaches the target node.

[0076] S23, pheromone update. After all robots complete path selection, they update the pheromones on the path. The pheromone update strategy is:

[0077] τ ij (t+1)=(1-ρ)τ ij (t)+Δτ ij (t,t+1)

[0078]

[0079] Among them, ρ is the pheromone volatility coefficient, Δτ ij is the pheromone increment during the robot's path-finding process, Q is the pheromone intensity, and L is the total path length of robot k.

[0080] S24, Path Evaluation. After each round of iteration, evaluate the path quality of all robots and record the best path currently found and the length of the best path. This process includes calculating the total length of each path, the cost involved in the path, the time efficiency, and other factors that may affect the quality of the path. Through these evaluations, the performance of each path can be quantified to understand their relative advantages and disadvantages.

[0081] After the evaluation is completed, the algorithm will record the optimal path and its corresponding path length within the iteration number, and use the optimal path and path length within the iteration number as a reference for the next round of iteration to help the algorithm optimize the search strategy. If a new optimal path is found, the algorithm will also update the record of the optimal path and adjust the distribution of pheromones to further strengthen the exploration of this path in the next iteration.

[0082] Repeat S21-S24 and iterate until the stopping condition is met, such as reaching the maximum number of iterations or finding a satisfactory path quality. The optimal path and the length corresponding to the optimal path are output as the final global path planning result.

[0083] S3: Establish a robot motion model. In local path planning, based on the robot's motion model, sample multiple speed combinations, perform trajectory analysis and prediction on each speed combination, evaluate the quality of the trajectory, and select the best speed group.

[0084] Figure 4 A flowchart of local optimal path planning provided by an embodiment of the present invention.

[0085] The steps to achieve local optimal path planning using the dynamic window method DWA are as follows: first, establish the robot kinematic model, calculate the robot's dynamic window, and determine the feasible speed range; second, evaluate the quality of the speed combination based on the target point and obstacle information; then, select the best speed combination to achieve the optimal path; finally, update the robot's position and adjust the dynamic window in real time, and continue path planning to ensure safe and efficient navigation.

[0086] DWA can sample the linear velocity v(t) and angular velocity w(t) of the mobile robot in the window, and establish the kinematic model of the mobile robot based on this. Assuming that the robot's trajectory can be decomposed into multiple time slices, and the robot moves in a uniform straight line in each time slice, the robot motion model is:

[0087]

[0088] Among them, v(t) is the robot's linear velocity, ω(t) is the robot's angular velocity, x(t) and y(t) represent the robot's position on the x-axis and y-axis at time t, θ(t) is the robot's orientation, and Δt is the time step. The range of variation of the linear velocity v(t) depends on the closest distance to the obstacle and the maximum linear deceleration; while the content of the angular velocity ω(t) is determined by both the closest distance to the obstacle and the maximum angular deceleration.

[0089] In velocity space, DWA describes the robot obstacle avoidance problem as a constrained optimization problem, including the incomplete motion constraints of the differential robot, the dynamic constraints of the robot structure, and the environmental constraints.

[0090] The velocity constraint of the robot is:

[0091] v s ={(v,ω)|v∈[v min ,v max ],ω∈[ω min ,ω max ]}.

[0092] Since different motors have different performances, the acceleration of the robot is also different. The acceleration constraint of the robot is:

[0093]

[0094] in, and are the minimum linear deceleration and minimum angular deceleration respectively; and are the maximum linear acceleration and maximum angular acceleration respectively.

[0095] The entire robot trajectory can be subdivided into several linear or circular motions. In order to ensure the safety of the robot, the robot is limited to completely avoid hitting obstacles under the maximum deceleration condition. Therefore, the robot's braking distance constraint is:

[0096]

[0097] Where dist(v,ω) is the distance between the simulated trajectory and the nearest obstacle.

[0098] After velocity space search and sampling, DWA generates the robot’s predicted motion trajectory within Ns based on the sampled multiple sets of velocity (v, ω) simulations, and then scores each trajectory according to the corresponding objective function.

[0099] The speed combination evaluation function is:

[0100] G(v,ω)=σ(A·heading(v,ω)+B·dist(v,ω)+C·velocity(v,ω))

[0101] Among them, G(v,ω) is the evaluation value of the simulated trajectory, heading(v,ω) represents the angle between the robot's heading and the target line, dist(v,ω) is the distance between the robot and the nearest obstacle, and velocity(v,ω) is used to evaluate the robot's motion performance. The greater the velocity, the higher the score. A, B, and C are the weight factors of the corresponding evaluation function; σ is the normalization factor.

[0102] From all the sampled velocity samples, select the motion instruction with the minimum cost. This instruction will be used as the next control command of the robot. Apply the selected velocity sample to the robot to update the velocity of the robot.

[0103] The DWA algorithm combines sensor data and motion models to evaluate the robot's motion state in real time and select the optimal motion instructions. This process is repeated continuously to cope with dynamically changing environments and achieve efficient local path planning. Through clear steps and feedback mechanisms, the DWA algorithm demonstrates its superiority in robot navigation and obstacle avoidance.

[0104] S4: In a multi-robot system, design conflict judgment rules to identify potential motion conflicts between robots; introduce a priority strategy to resolve conflicts based on the importance and urgency of the robot tasks.

[0105] The conflict judgment rule design process specifically includes the following steps:

[0106] The path planning problem of a single mobile robot is expanded to the path planning problem of multiple mobile robots. Based on the improved ACO algorithm, the global optimal path is pre-planned for each mobile robot in a grid environment. The DWA obstacle avoidance strategy is mainly applicable to static environments or local small-scale dynamic environments (low-speed motion with obstacle interference, etc.), but it is less effective in solving the motion conflict problem between multiple robots. The current mainstream coordinated collision avoidance technology can dissipate local conflicts and achieve collision avoidance coordination. In order to achieve better results in global multi-robot motion planning, the present invention combines the improved ACO, the improved DWA strategy and the multi-robot priority strategy to simplify the path planning problem into a dynamic path planning problem with order, thereby simplifying the complexity of the algorithm.

[0107] S41, setting a corresponding priority for each robot. When a low-priority robot encounters a high-priority robot during movement and a movement conflict is about to occur, the low-priority robot stops moving, and the high-priority robot performs obstacle avoidance movement based on DWA.

[0108] S42: When the actual distance between the two robots is less than the expected obstacle avoidance distance, there is a potential risk of collision. Use the low-priority robot to establish a local coordinate axis, calculate the angle α between the straight line of the two robots and the positive direction of the x-axis, and then calculate the difference between α and the orientation angle β of the high-priority robot.

[0109] S43, when |α-β|<90°, the risk of collision is extremely high. In order to significantly ensure the safety of multi-robot movement, the low-priority robot decelerates and stops in a short time, and the high-priority robot regards the low-priority robot as an unknown static obstacle and performs obstacle avoidance movement through DWA.

[0110] Therefore, the distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA in the present invention uses the improved dynamic window method (DWA) and ant colony optimization (ACO) algorithm to achieve effective local path planning and global path optimization in an environment with high-density obstacles and unknown static obstacles, ensuring the safe obstacle avoidance of robots. At the same time, motion conflict judgment rules and priority strategies in multi-robot systems are proposed to coordinate the motion between robots and obstacles to ensure safe and efficient navigation.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA, characterized in that: The following steps are involved: S1: Use sensors to obtain global environmental information and build an environmental model; the global environmental information includes static obstacle distribution and dynamic obstacle distribution; S2: Use the ant colony algorithm ACO for global path planning to optimize path length and number of turns; S3: Establish a robot motion model. In local path planning, based on the robot motion model, sample multiple speed combinations, analyze and predict the trajectory of each speed combination, evaluate the quality of the trajectory, and select the best speed group. S4: In a multi-robot system, design conflict judgment rules to identify potential motion conflicts between robots; introduce a priority strategy to resolve conflicts based on the importance and urgency of the robot tasks.

2. The distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA according to claim 1, characterized in that: The S1 specifically includes the following steps: S11, using a grid method to divide the entire free space into grid cells of uniform size to form a matrix MAP; S12, marking the matrix MAP with environmental information according to the acquired global environmental information; S13, marking a serial number for each grid cell in the matrix MAP; S14. The robot obtains environmental information through sensors and dynamically updates the environmental information of each grid unit.

3. The distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA according to claim 2, characterized in that: The S2 specifically includes the following steps: S21. Improve the initial pheromone distribution and allocate pheromone concentration values ​​on the grid cells according to the percentage of obstacles around the grid cell nodes. t ij (0)=ε*f(j) Among them, τ ij is the pheromone concentration value on the path from node i to node j, ε is the enhancement factor of the initial pheromone, f(j) represents the obstacle avoidance factor, C U is the complementary set symbol, allowed j Represents the number of obstacle grids around point j; the fewer the number of obstacle grids in the surrounding grid cells, the more significant the initial pheromone of the grid; S22. The probability of the robot selecting the next node is: in, is the probability that robot k moves from node i to node j at time t; η ij (t) is the heuristic function; the robot randomly selects a path based on the selection probability and performs global path planning; S23. After all robots complete path selection, they update the pheromones on the path. The pheromone update strategy is: t ij (t+1)=(1-ρ)τ ij (t)+Δτ ij (t,t+1) Among them, ρ is the pheromone volatility coefficient, Δτ ij is the pheromone increment during the robot's path-finding process, Q is the pheromone intensity, and L is the total path length of robot k; S24. Compare the path qualities of all robots, obtain the optimal path and its corresponding path length within the iteration number, and use the optimal path and path length within the iteration number as a reference for the next round of iteration to optimize the path; after the stopping condition is met, output the optimal path and the length corresponding to the optimal path as the final global path planning result.

4. The distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA according to claim 3 is characterized by: In S3, the robot motion model is Where v(t) is the linear velocity of the robot, ω(t) is the angular velocity of the robot, x(t) and y(t) represent the positions of the robot on the x-axis and y-axis at time t, respectively, θ(t) is the orientation of the robot, and Δt is the time step.

5. The distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA according to claim 4, characterized in that: In S3, the robot's velocity constraint is: v s ={(v,ω)|v∈[v min ,v max ],ω∈[ω min ,ω max ]}。 6. The distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA according to claim 5, characterized in that: In S3, the acceleration constraint of the robot is: in, and are the minimum linear deceleration and minimum angular deceleration respectively; and are the maximum linear acceleration and maximum angular acceleration respectively.

7. The distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA according to claim 6, characterized in that: In S3, the braking distance constraint of the robot is: Where dist(v,ω) is the distance between the simulated trajectory and the nearest obstacle.

8. The distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA according to claim 7, characterized in that: In S3, the speed combination evaluation function is: G(v,ω)=σ(A·heading(v,ω)+B·dist(v,ω)+C·velocity(v,ω)) Among them, G(v,ω) is the evaluation value of the simulated trajectory, heading(v,ω) represents the angle between the robot’s heading and the target line, dist(v,ω) is the distance between the robot and the nearest obstacle, velocity(v,ω) is used to evaluate the robot’s motion performance, A, B, and C are the weight factors of the corresponding evaluation functions, and σ is the normalization factor.

9. The distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA according to claim 8, characterized in that: In S4, the conflict judgment rule design process specifically includes the following steps: S41. Set a corresponding priority for each robot; S42. When the actual distance between the two robots is less than the expected obstacle avoidance distance, the low-priority robot establishes a local coordinate axis, calculates the angle α between the straight line of the two robots and the positive direction of the x-axis, and then calculates the difference between α and the orientation angle β of the high-priority robot. S43, when |α-β<90°, the low-priority robot decelerates and stops in a short time, and the high-priority robot regards the low-priority robot as an unknown static obstacle and performs obstacle avoidance movement through DWA.

Images