A Local Path Planning Method for Mobile Robots Based on Parameter Adaptive Dynamic Window Approach

Abstract

The present invention discloses a local path planning method for a mobile robot based on a parameter adaptive dynamic window method. Aiming at the problems existing in the local path planning of a mobile robot by the traditional dynamic window method, such as the robot colliding with obstacles due to improper setting of the parameters of the trajectory evaluation function and being unable to reach the local target point. The present invention proposes a parameter adaptive dynamic window method, comprehensively considering the robot performance and environmental obstacle information, optimizing the obstacle avoidance evaluation function and the weight factors of each evaluation index in the trajectory evaluation function, realizing the adaptive adjustment of the weight of each evaluation index during the local path planning process of the mobile robot, ensuring that the robot safely reaches the local target point, and at the same time optimizing the running path.

Description

Technical Field

[0001] The present invention relates to the technical field of mobile robot control, and in particular to a local path planning method for mobile robots based on a parameter adaptive dynamic window method. Background Art

[0002] Mobile robot path planning technology can be divided into global path planning focusing on path optimization and local path planning focusing on safe obstacle avoidance. Local path planning belongs to online planning, and sensors need to collect environmental information in real time to dynamically correct the orientation, enabling the mobile robot to have good real-time obstacle avoidance capabilities. Currently, common methods for local path planning technology include the dynamic window method based on the dynamic model, the neural network algorithm based on biological intelligence, the artificial potential field method based on the virtual potential field function, etc.

[0003] The dynamic window method fully considers the mechanical characteristic limitations of the mobile robot itself and the constraints of the obstacle environment on the running speed of the mobile robot. It forms a dynamic window for speed sampling according to the constraints, samples multiple groups of speeds, and simulates the movement trajectories of the mobile robot within the next sampling time at these speeds. Then, through the trajectory evaluation function, these trajectories are evaluated and processed, and the sampling speed corresponding to the trajectory with the highest score is selected as the optimal speed group to drive the mobile robot towards the local target point. The path planned using the dynamic window method can achieve safe obstacle avoidance of local environment obstacles by the mobile robot, and the algorithm has small computational complexity, rapid response, and high reliability. Therefore, it has been widely applied in the local path planning technology of mobile robots.

[0004] However, in the application of the local path planning field of mobile robots, the traditional dynamic window method has the following deficiencies:

[0005] (1) The weight factor α of the guidance evaluation function is a constant set based on prior knowledge. If it is set too large, the planned path is prone to overly tend towards the local target point and fail to complete local obstacle avoidance. If it is set too small, it cannot successfully reach the local target point.

[0006] (2) The weight factor β of the obstacle avoidance evaluation function is a constant set based on prior knowledge. If it is set too large, the robot is prone to move away from the obstacle due to excessive obstacle avoidance, resulting in a significant increase in path cost and running time. If it is set too small, it is prone to collide with the obstacle, affecting the safe movement of the mobile robot.

[0007] (3) The weight factor γ of the speed evaluation function is a constant set based on prior knowledge. If it is set too large, the robot is prone to collide with the obstacle due to excessive speed. If it is set too small, it is prone to stagnate at the equilibrium point and cannot reach the local target point. Summary of the Invention

[0008] To address the deficiencies of the above traditional dynamic window method, the main objective of the present invention is to propose a local path planning method for mobile robots based on a parameter - adaptive dynamic window method, in order to solve the problems that when using the traditional dynamic window method for local path planning of mobile robots, local obstacle avoidance cannot be completed and the local target point cannot be successfully reached due to improper setting of the trajectory evaluation function parameters.

[0009] To achieve the above objective, the present invention provides a local path planning method for mobile robots based on a parameter - adaptive dynamic window method. When the traditional dynamic window method is used for local path planning of mobile robots, the setting of the weight factor of the guidance evaluation function depends on the previous experience of experimenters. If it is set too large, the planned path may not be able to complete local obstacle avoidance because it tends too much towards the local target point. If it is set too small, it may not be able to accurately complete the target guidance and thus cannot successfully reach the local target point. The present invention introduces an adaptive parameter ζ1 to optimize the weight factor of the guidance evaluation function. According to the deviation between the heading angle of the mobile robot at the predicted position and the desired heading angle, the weight factor of the guidance evaluation function is adaptively adjusted to correct the heading angle of the mobile robot in real - time, ensuring that the robot safely reaches the local target point.

[0010] Furthermore, the parameter - adaptive dynamic window method proposed in the present invention optimizes the definition formula of the obstacle - avoidance evaluation function and introduces an adaptive parameter ζ2 to optimize the weight factor of the obstacle - avoidance evaluation function. According to the real - time distance between the mobile robot and the obstacle during the movement process, the size of the weight factor of the obstacle - avoidance evaluation function is adaptively adjusted, improving the safety performance of the running path of the mobile robot in a complex environment. At the same time, it avoids the planned path being too far away from the obstacle due to excessive obstacle avoidance, resulting in a significant increase in the path length cost and running time, and effectively reducing the mechanical energy consumption of the mobile robot.

[0011] Finally, an adaptive parameter ζ3 is introduced to optimize the weight factor of the speed evaluation function in the trajectory evaluation function, incorporating information on static and dynamic obstacles in the environment. The weight of the speed evaluation function is adaptively adjusted according to the complexity of the operating environment, achieving efficient local obstacle avoidance and successful reaching of the local target point by the mobile robot on the premise of ensuring path safety, and improving the operating efficiency of the algorithm.

[0012] The technical solution adopted by the present invention is as follows:

[0013] First, it is stated that the following parameter - adaptive dynamic window method is based on the following assumptions:

[0014] 1. It is assumed that the local map and the obstacle boundary are established considering the safety distance of the mobile robot's volume. Therefore, the movement of the mobile robot can be regarded as the movement of a mass point.

[0015] 2. The present invention simplifies the working environment of the mobile robot into a two-dimensional local grid map, simplifies the obstacles in the environment into circular obstacles in the grid map, and only considers the movement and position change of the mobile robot in the local grid map;

[0016] 3. The motion model of the mobile robot in the present invention is shown as follows:

[0017]

[0018] A local path planning method for a mobile robot based on a parameter adaptive dynamic window method mainly includes the following steps:

[0019] Step 1: Establish a local grid map, and determine the positions of the local starting point, local target point, and obstacles;

[0020] Step 2: Use the traditional dynamic window method for preliminary path planning to provide an algorithm platform for the local path planning of the mobile robot based on the parameter adaptive dynamic window method;

[0021] Step 3: Optimize the weight factor of the guidance evaluation function in the trajectory evaluation function of the traditional dynamic window method. According to the deviation between the predicted position heading angle and the desired heading angle of the mobile robot, adaptively adjust the weight factor of the guidance evaluation function, and real-time correct the heading angle of the robot to ensure that the robot safely reaches the local target point;

[0022] Step 4: Optimize the obstacle avoidance evaluation function and its weight factor in the trajectory evaluation function of the traditional dynamic window method. According to the real-time distance between the mobile robot and the obstacles during the movement process, adaptively adjust the size of the weight factor of the obstacle avoidance evaluation function, improve the safety performance of the running path of the mobile robot in a complex environment, and at the same time avoid the planned path being too far away from the obstacles due to excessive obstacle avoidance, resulting in a large increase in the path length cost and running time, and effectively reducing the mechanical energy consumption of the mobile robot;

[0023] Step 5: Optimize the weight factor of the speed evaluation function in the trajectory evaluation function of the traditional dynamic window method. Incorporate the information of static and dynamic obstacles in the environment, adaptively adjust the weight of the speed evaluation function according to the complexity of the running environment, and realize the mobile robot to efficiently complete local obstacle avoidance and reach the local target point on the premise of ensuring path safety, and improve the running efficiency of the algorithm.

[0024] Furthermore, in order to complete the local obstacle avoidance of the mobile robot and safely reach the local target point, the present invention optimizes the weight factor of the guidance evaluation function in the trajectory evaluation function in Step 3. The optimized weight factor of the guidance evaluation function is ζ1, and the definition formula is as follows:

[0025] ζ1 = kα

[0026] Wherein, the value of k is shown as follows:

[0027]

[0028] Further, ψ is the guiding angle, which represents the heading angle of the predicted position of the mobile robot at time t. And the desired heading angle The deviation amount between them. The definition formula is:

[0029]

[0030] Among them, the desired heading angle Represents the angle between the line connecting the predicted position of the robot to the local target point and the positive direction of the X-axis in the earth coordinate system XOY. The calculation formula is as follows:

[0031]

[0032] Among them, (x tp , y tp ) are the coordinates of the local target point, Is the coordinate information of the predicted position of the mobile robot.

[0033] It can be seen from the definition of ζ1 that when Is satisfied, the guiding angle of the predicted trajectory is smaller at this time, that is, the deviation between the heading angle of the predicted position of the robot and the desired heading angle is smaller. The robot will continuously approach along the direction of the target point. Therefore, a reward mechanism is given to such predicted trajectories, and the weight of the guidance evaluation function in the trajectory evaluation function is increased, so as to prompt the robot to move quickly in the target direction and reach the target point, which is beneficial to improving the operation efficiency of the algorithm and reducing the mechanical energy consumption of the robot. When Is satisfied, the guiding angle of the predicted trajectory is larger at this time, that is, the deviation between the heading angle of the predicted position of the robot and the desired heading angle is larger. The robot will move along the direction deviating from the local target point. Therefore, a penalty mechanism is given to such predicted trajectories, and the weight of the guidance evaluation function in the trajectory evaluation function is reduced, so as to prompt the mobile robot to correct the heading angle.

[0034] Further, to improve the safety performance of the local path planning of the mobile robot in a complex dynamic environment, the present invention improves the definition formula of the obstacle avoidance evaluation function as follows: First, introduce the static obstacle set P1 on the corresponding predicted trajectory curvature and the dynamic obstacle set P2 on the corresponding predicted trajectory curvature, which respectively satisfy:

[0035]

[0036]

[0037] At the same time, define the optimized obstacle avoidance evaluation function as dist1(v, ω), and the expression is discussed in the following four cases:

[0038] (1) If then take dist1(v, ω) = 4R safe ;

[0039] (2) If then take dist1(v, ω) = dist_s(v, ω);

[0040] (3) If then take dist1(v, ω) = dist_d(v, ω);

[0041] (4) If then dist1(v, ω) = min{dist_s(v, ω), dist_d(v, ω)};

[0042] Among them, dist_s(v, ω) and dist_d(v, ω) are the static obstacle evaluation function and the dynamic obstacle evaluation function respectively, and R safe represents the safety distance of the mobile robot, and the definition formula is as follows:

[0043] R safe = 1.2R obs

[0044] Among them, R obs is the radius of the circular obstacle in the environment.

[0045] Furthermore, to further optimize the local obstacle avoidance ability of the mobile robot and avoid the problem of too large obstacle avoidance evaluation function value when the distance between the robot and the obstacle on the corresponding predicted trajectory curvature is large, the parameter adaptive dynamic window method defines the static obstacle evaluation function dist_s(v, ω) and the dynamic obstacle evaluation function dist_d(v, ω) as follows:

[0046]

[0047]

[0048] Among them, represents the distance between the mobile robot and the nearest static obstacle on the corresponding predicted trajectory curvature, represents the distance between the mobile robot and the nearest dynamic obstacle on the corresponding predicted trajectory curvature. The calculation formulas are shown as follows respectively:

[0049]

[0050]

[0051] Among them, (x t , y t) is the position coordinate of the mobile robot at time t, represents the position coordinate of the i-th static obstacle, represents the position coordinate of the j-th dynamic obstacle.

[0052] Furthermore, in order to improve the safety performance of the operation path of the mobile robot in a complex environment, and at the same time avoid the planned path being far away from obstacles due to excessive obstacle avoidance, resulting in a significant increase in the path length cost and operation time, the present invention optimizes the weight factor of the obstacle avoidance evaluation function in the trajectory evaluation function in step 4. The optimized weight factor of the obstacle avoidance evaluation function is ζ2, and the definition formula is as follows:

[0053]

[0054] It can be seen from the definition of ζ2 that when the mobile robot is close to the obstacle, the weight of the obstacle avoidance evaluation function in the trajectory evaluation function is increased, prompting the robot to preferentially select a predicted trajectory farther from the obstacle and bypass the obstacle to complete local safe obstacle avoidance. When the mobile robot is far from the obstacle, the probability of the predicted trajectory of the robot colliding with the obstacle is extremely small at this time. Therefore, the weight of the obstacle avoidance evaluation function in the trajectory evaluation function is reduced to avoid the influence of the obstacle avoidance evaluation function on the trajectory evaluation function being too large, thereby improving the operation efficiency of the algorithm and reducing the energy loss of the robot.

[0055] Furthermore, to avoid the mobile robot colliding with obstacles in the environment due to excessive speed and at the same time improve the operation efficiency of the algorithm, the present invention optimizes the weight factor of the speed evaluation function in the trajectory evaluation function in step 5. Define the optimized weight factor of the speed evaluation function as ζ3, and the definition formula is as follows:

[0056]

[0057] where, H is the occupancy ratio of static obstacles in the local path planning area of the mobile robot, and τ(t) is the collision risk factor between the mobile robot and the nearest dynamic obstacle on the curvature of the corresponding predicted trajectory.

[0058] Furthermore, considering the influence of static obstacles in the planning area on the local path planning of the mobile robot, in order to improve the safety of the local planned path of the mobile robot and the operation efficiency of the algorithm, the present invention introduces the occupancy ratio of static obstacles in the local planning area, which is represented by H, and the definition formula is as follows:

[0059]

[0060] where, N represents the number of grid cells of static obstacles in a rectangular area with the task start point and the local target point as two diagonal vertices, (x sp ,y sp) represents the starting point of the local path planning task, (x tp , y tp ) represents the target point of the local path planning task, and [·] represents the rounding function.

[0061] It can be seen from the definition of ζ3 that the more static obstacles there are in the local environment, the larger the value of N, the larger the value of H, and the smaller the value of arccosH. As a result, the weight of the speed evaluation function in the trajectory evaluation function is reduced, avoiding the robot from colliding with static obstacles in the environment due to excessive speed and improving the safety performance of the local path. When there are fewer static obstacles in the local environment, the values of N and H decrease, then the value of arccosH and the parameter ζ3 value increase, thereby increasing the weight of the speed evaluation function in the trajectory evaluation function, increasing the movement speed of the robot, and prompting the robot to quickly reach the local target point on the premise of ensuring the path safety performance, improving the operation efficiency of the algorithm.

[0062] Furthermore, in a dynamic environment, considering the influence of dynamic obstacles on the local path planning of mobile robots, to further improve the safety of the local planned path of mobile robots and effectively achieve dynamic real-time obstacle avoidance. The parameter adaptive dynamic window method in the present invention introduces a collision risk factor τ(t), and the definition formula is:

[0063]

[0064] where, ||.|| represents the norm, v ro (t) represents the relative speed of the mobile robot and the nearest dynamic obstacle on the corresponding predicted trajectory curvature at time t, that is

[0065]

[0066] Furthermore, represents the collision angle between the mobile robot and the nearest dynamic obstacle on the corresponding predicted trajectory curvature at time t, and its definition formula is:

[0067]

[0068] where, represents the relative position vector between the robot and the nearest dynamic obstacle on the corresponding predicted trajectory curvature at time t, that is

[0069]

[0070] where, is the state information of the nearest dynamic obstacle on the corresponding predicted trajectory curvature to the mobile robot at time t.

[0071] It can be seen from the definition of ζ3 that at time t, the collision risk factor τ(t) of the robot increases with the collision angle of the robot It decreases with the increase of . When the collision angle increases, the robot moves away from the dynamic obstacle, the value of τ(t) decreases, and the value of parameter ζ3 increases, thereby increasing the weight of the velocity evaluation function in the trajectory evaluation function, increasing the movement speed of the robot, and enabling the robot to quickly reach the local target point on the premise of ensuring the path safety performance, improving the algorithm efficiency. When the collision angle is small, the robot moves towards the dynamic obstacle, the value of τ(t) increases, and the value of parameter ζ3 decreases, thereby reducing the weight of the velocity evaluation function in the trajectory evaluation function, reducing the movement speed of the robot, and avoiding the robot colliding with the dynamic obstacle in the environment due to excessive speed, improving the safety performance of the local path. If there is no dynamic obstacle on the corresponding predicted trajectory curvature, the collision risk factor τ(t) takes a value of 0.

[0072] Furthermore, based on the above improvements, when the parameter adaptive dynamic window method in the present invention evaluates the predicted trajectory, the trajectory evaluation function is optimized, and the optimized expression is as follows:

[0073] G1(v, ω) = σ(ζ1angle(v, ω) + ζ2dist1(v, ω) + ζ3velocity(v, ω))

[0074] Among them, angle(v, ω) represents the guidance evaluation function, dist1(v, ω) represents the obstacle avoidance evaluation function, velocity(v, ω) represents the velocity evaluation function, σ is the normalization factor, and ζ1, ζ2, ζ3 are the weight factors of each evaluation index.

[0075] The present invention adopts the above technical solutions. In the local path planning of a mobile robot, the planned path can safely avoid obstacles, reduce the path length cost, improve the operation efficiency, and successfully reach the local target point. This shows that the method in the present invention can not only improve the safety performance of the movement of the mobile robot, but also effectively reduce the mechanical energy consumption of the mobile robot. Brief Description of the Drawings

[0076] The following further describes the present invention in detail in conjunction with the drawings and specific embodiments;

[0077] Figure 1 It is a flow chart of the local path planning method for a mobile robot based on the parameter adaptive dynamic window method of the present invention;

[0078] Figure 2 It is a simulation diagram of the local path planning result of a mobile robot based on the traditional dynamic window method;

[0079] Figure 3 It is a simulation diagram of the local path planning result of a mobile robot based on the parameter adaptive dynamic window method of the present invention;

[0080] Figure 4 Comparison table of simulation results of local path planning of mobile robots before and after optimization by dynamic window method Specific implementation mode

[0081] In order to make the improvement objectives, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the following embodiments are only a part of the embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present invention.

[0082] Currently, in the application of the local path planning field of mobile robots, the traditional dynamic window method has the following deficiencies: the weight factors of each evaluation index in the trajectory evaluation function are constants set based on the previous experience of experimental personnel, so the algorithm has poor robustness and poor adaptability to the local path planning tasks of mobile robots in complex environments. Based on this, the main purpose of this application is to propose a local path planning method for mobile robots based on the parameter adaptive dynamic window method. By optimizing the parameters in the trajectory evaluation function and combining the robot's own performance and environmental information, the parameters in the trajectory evaluation function are adaptively adjusted, so as to improve the safety performance and operation efficiency of the local path planning of mobile robots.

[0083] For the convenience of understanding this embodiment, first, a local path planning method for mobile robots based on the parameter adaptive dynamic window method disclosed in the embodiments of the present invention will be introduced in detail.

[0084] Embodiment:

[0085] A local path planning method for mobile robots based on the parameter adaptive dynamic window method mainly includes the following steps:

[0086] Step 1, establish a local grid map, and determine the positions of the task starting point, local target point and obstacles;

[0087] Step 2, use the traditional dynamic window method for preliminary path planning to provide an algorithm platform for the local path planning of mobile robots based on the parameter adaptive dynamic window method;

[0088] Step 3, optimize the weight factor of the guidance evaluation function in the trajectory evaluation function of the traditional dynamic window method, and adaptively adjust the weight factor of the guidance evaluation function according to the deviation between the predicted position heading angle and the desired heading angle of the mobile robot, and correct the heading angle of the robot in real time to ensure that the mobile robot safely reaches the local target point;

[0089] Step 4: Optimize the obstacle avoidance evaluation function and its weight factor in the trajectory evaluation function of the traditional dynamic window method. According to the real-time distance between the mobile robot and the obstacles during the movement process, adaptively adjust the size of the weight factor of the obstacle avoidance evaluation function, improve the safety performance of the running path of the mobile robot in a complex environment, and at the same time avoid the planned path being too far away from the obstacles due to excessive obstacle avoidance, resulting in a significant increase in the path length cost and running time, and effectively reduce the mechanical energy consumption of the mobile robot.

[0090] Step 5: Optimize the weight factor of the speed evaluation function in the trajectory evaluation function of the traditional dynamic window method, incorporate the information of static and dynamic obstacles in the environment, and adaptively adjust the weight of the speed evaluation function according to the complexity of the running environment, so as to achieve the efficient completion of local obstacle avoidance and reaching the local target point by the mobile robot on the premise of ensuring path safety, and improve the running efficiency of the algorithm.

[0091] Furthermore, in Step 1, use the matrix method to establish a local grid map on the MATLAB R2018a simulation platform, and determine the local starting point, local target point and the positions of the obstacles for the local path planning task of the mobile robot. Among them, the coordinates of the local starting point are (0, 0), and the coordinates of the local target point are (10, 10).

[0092] Furthermore, in Step 2, call the traditional dynamic window method for local path planning of the mobile robot, and the simulation results are as attached Figure 2 shown. The results show that when the parameter values in the trajectory evaluation function are α = 10, β = 0.5, γ = 0.5, the weight factor of the guidance function is set too large, and the movement path of the robot moves along the direction pointing to the local target point and cannot complete local obstacle avoidance. The simulation results are as attached Figure 2 (a) shown. When the parameter values in the trajectory evaluation function are α = 0.01, β = 0.5, γ = 0.5, the weight factor of the guidance function is set too small, and the movement path of the mobile robot does not move along the direction of the local target point and cannot reach the local target point successfully. The simulation results are as attached Figure 2 (b) shown. When the parameter values in the trajectory evaluation function are α = 0.5, β = 10, γ = 0.5, the weight factor of the obstacle avoidance function is set too large. Although the robot can reach the local target point successfully, the smoothness of the movement path is poor and the running time is long due to excessive obstacle avoidance. The simulation results are as attached Figure 2 (c) shown. When the parameter values in the trajectory evaluation function are α = 0.5, β = 0.01, γ = 0.5, the weight factor of the obstacle avoidance function is set too small, and the robot collides with the obstacles in the environment, resulting in the failure of path planning. The simulation results are as attached Figure 2(as shown in (d)). When the parameter values in the trajectory evaluation function are α = 0.5, β = 0.5, and γ = 10, the weight factor of the velocity function is set too large, and the robot collides with the obstacle due to excessive movement speed. The simulation results are as attached Figure 2 (e). When the parameter values in the trajectory evaluation function are α = 0.5, β = 0.5, and γ = 0.01, the weight factor of the velocity function is set too small, and the robot stagnates at the local equilibrium point due to too slow movement speed and cannot reach the local target point successfully. The simulation results are as attached Figure 2 (f).

[0093] Furthermore, in step 3, the optimized guidance evaluation function weight factor is defined as ζ1, and the definition formula is as follows:

[0094] ζ1 = kα

[0095] The value of k in the formula is shown as follows:

[0096]

[0097] Among them, the definition formula of the guiding angle ψ is: Desired heading angle is the angle between the line connecting the predicted position of the robot to the local target point and the positive direction of the X-axis in the earth coordinate system XOY.

[0098] Furthermore, introduce the static obstacle set P1 and the dynamic obstacle set P2 on the corresponding predicted trajectory curvature, and define the optimized obstacle avoidance evaluation function as dist1(v, ω). The expression is discussed in the following four cases:

[0099] (1) If then take dist1(v, ω) = 4R safe ;

[0100] (2) If then take dist1(v, ω) = dist_s(v, ω);

[0101] (3) If then take dist1(v, ω) = dist_d(v, ω);

[0102] (4) If then dist1(v, ω) = min{dist_s(v, ω), dist_d(v, ω)};

[0103] Among them, dist_s(v, ω) and dist_d(v, ω) are the static obstacle evaluation function and the dynamic obstacle evaluation function respectively, and the definition formulas are as follows:

[0104]

[0105]

[0106] Among them, represents the distance between the mobile robot and the nearest static obstacle on the corresponding predicted trajectory curvature, represents the distance between the mobile robot and the nearest dynamic obstacle on the corresponding predicted trajectory curvature. R safe represents the safety distance of the mobile robot, and is calculated by R safe = 1.2R obs where R obs is the radius of the circular obstacle in the environment.

[0107] Furthermore, in step 4, the weight factor of the optimized obstacle avoidance evaluation function is defined as ζ2, and the definition formula is as follows:

[0108]

[0109] Furthermore, in step 5, the weight factor of the optimized speed evaluation function is defined as ζ3, and the definition formula is as follows:

[0110]

[0111] where H is the occupancy ratio of static obstacles in the local path planning area of the mobile robot, and the calculation formula is as follows:

[0112]

[0113] where N represents the number of grid cells of static obstacles in the rectangular area with the task start point and the local target point as two diagonal vertices, (x sp , y sp ) is the start point of the local path planning task of the mobile robot, (x tp , y tp ) is the target point of the local path planning task of the mobile robot, and [·] is the rounding function.

[0114] Furthermore, τ(t) is the collision risk factor between the mobile robot and the nearest dynamic obstacle on the corresponding predicted trajectory curvature, and the definition formula is:

[0115]

[0116] where ||.|| represents the norm, v ro (t) is the relative velocity between the mobile robot and the nearest dynamic obstacle on the corresponding predicted trajectory curvature at time t, is the collision angle between the mobile robot and the nearest dynamic obstacle on the corresponding predicted trajectory curvature at time t, and is calculated by ​ It is the relative position vector between the robot and the nearest dynamic obstacle on the corresponding predicted trajectory curvature at time t. If there is no dynamic obstacle on the corresponding predicted trajectory curvature, the collision risk factor τ(t) takes the value of 0.

[0117] Furthermore, based on the above improvements, the parameter adaptive dynamic window method in the present invention optimizes the parameters of the trajectory evaluation function in the traditional dynamic window method, and the optimized expression is as follows:

[0118] G1(v,ω)=σ(ζ1angle(v,ω)+ζ2dist1(v,ω)+ζ3velocity(v,ω))

[0119] Among them, angle(v,ω) represents the guidance evaluation function, dist1(v,ω) represents the obstacle avoidance evaluation function, velocity(v,ω) represents the speed evaluation function, σ is the normalization factor, and ζ1, ζ2, ζ3 are the weight factors of each evaluation index.

[0120] Furthermore, based on the parameter adaptive dynamic window method in the present invention, the simulation results of the local path planning of the mobile robot are as shown in the appendix Figure 3 As shown. The results show that compared with the traditional dynamic window method, the path planned by the parameter adaptive dynamic window method in the present invention improves the smoothness of the path on the premise of ensuring that the mobile robot completes local obstacle avoidance, runs safely and reaches the local target point successfully, and effectively reduces the mechanical energy consumption of the robot. At the same time, the running time of the algorithm is reduced by 85.8%, greatly improving the running efficiency of the algorithm, which is of great significance for improving the economic benefits in social production practice.

[0121] The comparison results of the local path planning simulation of the mobile robot based on the traditional dynamic window method and the parameter adaptive dynamic window method in this embodiment are as shown in the appendix Figure 4 As shown.

Claims

1. A local path planning method for a mobile robot based on a parameter adaptive dynamic window approach, characterized by According to the real-time distance between the mobile robot and obstacles during movement, adaptively adjust the size of the weight factor of the obstacle avoidance evaluation function. Define the weight factor of the obstacle avoidance evaluation function in the trajectory evaluation function as ζ2, and the definition formula is as follows: Among them, dist1(v, ω) is the obstacle avoidance evaluation function, and R safe is the safety distance of the mobile robot.

2. The local path planning method of the mobile robot according to claim 1, wherein To further improve the local obstacle avoidance ability of the mobile robot and avoid the problem of too large an obstacle avoidance evaluation function value when the distance between the robot and the obstacles on the corresponding predicted trajectory curvature is far, introduce the static obstacle set P1 and the dynamic obstacle set P2 on the corresponding predicted trajectory curvature to optimize the obstacle avoidance evaluation function in the trajectory evaluation function. Define the optimized obstacle avoidance evaluation function as dist1(v,ω), and the expression is discussed in the following four cases: (1) If then take dist1(v, ω) = 4R safe ; (2) If then take dist1(v, ω) = dist_s(v, ω); (3) If then take dist1(v, ω) = dist_d(v, ω); (4) If then dist1(v, ω) = min{dist_s(v, ω), dist_d(v, ω)}; Among them, dist_s(v, ω) and dist_d(v, ω) are the static obstacle evaluation function and the dynamic obstacle evaluation function respectively, and R safe represents the safety distance of the mobile robot, and the definitions are as follows: R safe = 1.2R obs wherein, represents the distance between the mobile robot and the nearest static obstacle on the corresponding predicted trajectory curvature, represents the distance between the mobile robot and the nearest dynamic obstacle on the corresponding predicted trajectory curvature, R obs is the radius of the circular obstacle.

3. A local path planning method for a mobile robot based on a parameter adaptive dynamic window method, characterized in that, Integrate the information of static and dynamic obstacles in the environment, and adaptively adjust the weight of the speed evaluation function according to the complexity of the operating environment. Define the weight factor of the speed evaluation function in the trajectory evaluation function as ζ3, and the definition formula is as follows: Among them, H is the occupancy ratio of static obstacles in the local path planning area of the mobile robot; τ(t) is the collision risk factor between the mobile robot and the nearest dynamic obstacle on the corresponding predicted trajectory curvature, and the definition formula is as follows: where, ||.|| is the norm, and v ro (t) is the relative velocity between the mobile robot and the nearest dynamic obstacle on the curvature of the corresponding predicted trajectory at time t, denotes the collision angle between the mobile robot and the nearest dynamic obstacle on the curvature of the corresponding predicted trajectory at time t.

Images