Local path planning method and system for mobile robots using adaptive parameters

Abstract

A method and system for local path planning for a mobile robot using adaptive parameters evaluates predicted trajectories using a trajectory evaluation model. When planning a path, the method traverses all predicted trajectories corresponding to multiple speed sets and determines the number of predicted trajectories that contain obstacles among all predicted trajectories. If the number of predicted trajectories that contain obstacles is greater than a set threshold, the method reduces the proportion of directional cost and the proportion of speed difference cost in the trajectory evaluation model and calculates a sampling speed based on the updated proportions. By adaptively adjusting the proportion of directional cost and the proportion of speed difference cost in the trajectory evaluation model, the problem of the device getting stuck in a local optimum can be effectively avoided without increasing the computational burden.

Description

[Technical Field]

[0001] (CROSS-REFERENCE TO RELATED APPLICATIONS) This invention claims priority to a Chinese patent application bearing application number 202310138722.2 and entitled "Method and system for local path planning for mobile robots with adaptive parameters," filed with the China Patent Office on February 21, 2023, the entire contents of which are incorporated herein by reference.

[0002] The present invention relates to the technical field of path optimization, and more particularly to a method and system for local path planning for a mobile robot with adaptive parameters. [Background technology]

[0003] Path planning algorithms are mainly divided into global path planning and local path planning. The Dynamic Window Approach (DWA) is a typical local path planner that tends to select a trajectory moving toward the goal direction. The DWA samples the linear and angular velocities in the velocity space, predicts the trajectory for the next time interval based on the kinematic model of the robot, and scores the trajectory to be evaluated, thereby obtaining a safer, smoother, and optimal local path.

[0004] The inventors have noticed that when using a conventional local path planning algorithm for a robot, when the robot is in a position where an obstacle is present ahead, the desired trajectory must not only move to the target, but also avoid the obstacle. This contradiction causes the robot to fall into a local minimum or local optimum, specifically, the robot will constantly go back and forth around this position, and may even stop, reducing the effectiveness of obstacle avoidance. Summary of the Invention

[0005] To solve the above problems, the present invention provides a method and system for local path planning for a mobile robot using adaptive parameters, which can prevent the robot from getting stuck in a local optimum during local path planning.

[0006] To achieve the above objectives, the present invention is realized by the following technical solutions.

[0007] In a first aspect, the present invention provides a method for producing a composition comprising: sampling in velocity space to obtain a plurality of velocity sets; performing trajectory prediction for each of the speed sets to obtain a predicted trajectory; evaluating the predicted trajectory using a preset trajectory evaluation model and adjusting parameters in the trajectory evaluation model to plan a route and obtain an optimal route; The trajectory evaluation model includes a direction cost, a speed difference cost, and a total trajectory cost, and all predicted trajectories corresponding to a plurality of speed sets are traversed to determine the number of predicted trajectories containing obstacles among all the predicted trajectories, and if the number of predicted trajectories containing obstacles is greater than a set threshold, the score proportions of the direction cost and the speed difference cost in the trajectory evaluation model are reduced, and a sampling speed is calculated based on the updated proportions.

[0008] Furthermore, the direction cost is the angle between the robot's current position and the local target position and the robot's current target direction, the velocity difference cost is the difference between the robot's current velocity and the robot's maximum velocity, and the total trajectory cost is the sum of the corresponding costs in the cost map of all trajectory points at a set of sampling velocities.

[0009] Furthermore, the trajectory evaluation model is a sum obtained by adding the product of the adaptation coefficient of the direction cost, the direction cost, and the score rate corresponding to the direction cost to the product of the adaptation coefficient of the speed difference cost, the speed difference cost, and the score rate corresponding to the speed difference cost, and further adding the product of the total trajectory cost value and the score rate corresponding to the total trajectory cost value.

[0010] Furthermore, the adaptive coefficient of the direction cost and the adaptive coefficient of the speed difference cost each take the maximum value between 0.1 and the predicted value, and the predicted value is the ratio of the difference between the total number of predicted trajectories and the number of trajectories where obstacles exist to the total number of predicted trajectories.

[0011] Furthermore, it is determined whether a collision will occur when performing an emergency stop for the trajectory point of the first control cycle on the predicted trajectory, and if so, the corresponding velocity sampling is discarded; otherwise, the corresponding costs in the cost map of the trajectory points in each control cycle are accumulated.

[0012] Furthermore, if the number of predicted trajectories containing obstacles is less than or equal to a set threshold, the robot's current point and the local target point are connected, and if obstacle information is present on the line connecting the robot's current point and the local target point, the path is re-planned by the global planner, the planned path is entered into the local planned target point set, the local target point is updated and substituted into the trajectory evaluation model, and the optimal sampling speed is calculated.

[0013] Furthermore, the global planner searches for route points to the destination, The Floyd path smoothing algorithm sets local path planning key points, Select the target point from the key points, As the device travels, it builds a local map and trajectory estimation model to adjust its local plan based on whether an obstacle exists between the current point and the destination point.

[0014] In a second aspect, the present invention provides a method for producing a pharmaceutical composition comprising: a sampling module configured to sample in velocity space to obtain a plurality of velocity sets; a prediction module configured to perform trajectory prediction for each of the speed sets to obtain a predicted trajectory; a planning module configured to evaluate a predicted trajectory according to a preset trajectory evaluation model, and to plan a route by adjusting parameters in the trajectory evaluation model to obtain an optimal route; The present invention further provides a local path planning system for a mobile robot using adaptive parameters, in which the trajectory evaluation model includes a direction cost, a speed difference cost, and a total trajectory cost, and traverses all predicted trajectories corresponding to a plurality of speed sets to determine the number of predicted trajectories containing obstacles among all the predicted trajectories, and if the number of predicted trajectories containing obstacles is greater than a set threshold, reduces the score proportion of the direction cost and the score proportion of the speed difference cost in the trajectory evaluation model, and calculates a sampling speed based on the updated proportions.

[0015] In a third aspect, the present invention further provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method for local path planning for a mobile robot using adaptive parameters as set forth in the first aspect.

[0016] In a fourth aspect, the present invention further provides an electronic device comprising: a memory; a processor; and a computer program stored in the memory and executable on the processor, the computer program implementing, when executed by the processor, the steps of the method for local path planning for a mobile robot using adaptive parameters as described in the first aspect.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0018] In this invention, when evaluating predicted trajectories using a trajectory evaluation model and planning a path, all predicted trajectories corresponding to multiple velocity sets are traversed to determine the number of predicted trajectories that contain obstacles among all predicted trajectories, and if the number of predicted trajectories that contain obstacles is greater than a set threshold, the proportions of directional cost and velocity difference cost in the trajectory evaluation model are reduced, and a sampling velocity is calculated based on the updated proportions. By adaptively adjusting the proportions of directional cost and velocity difference cost in the trajectory evaluation model, the problem of the robot falling into a local optimum can be effectively avoided without increasing the computational burden.

[0019] The drawings in the specification that form a part of this embodiment are intended to further understand this embodiment, and the exemplary embodiments and descriptions thereof are intended to interpret this embodiment and are not intended to unduly limit this embodiment. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a flowchart of a first embodiment of the present invention. [Figure 2] This is a solution of the local planner in the first embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0021] The invention will now be further described with reference to the following figures and examples.

[0022] It should be noted that the following detailed description is for illustrative purposes only and is intended to further explain the present application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art.

[0023] Example 1 This embodiment provides a local path planning method for a mobile robot with adaptive parameters, including:

[0024] S1: Search for route points to the destination point using the global planner.

[0025] S2. Using the Floyd path smoothing algorithm, the path points that are on a straight line from the starting point and exceed a certain threshold are set as key points for local path planning. It can be understood that there are no obstacles on this straight line, where the straight line refers to the line connecting the current point and the local destination point, and the threshold can be understood as the shortest distance between the local path key points.

[0026] S3: Select a local target point to be tracked based on the current position of the robot and the threshold value of the target point to be tracked, and select the target point from the key points. Specifically, the selection process must satisfy the condition that the distance from the current point exceeds a certain threshold and the point is selected as the local target point for the first time.

[0027] S4: When the robot is traveling, it adjusts the local plan based on whether there is an obstacle between the current point and the local goal point by constructing a local map and a sampling evaluation model. Specifically, when constructing a local map, this embodiment chooses to first delete the data information of the previous frame, and then construct the obstacle information based on the data of the current frame. The sampling evaluation model is a trajectory evaluation model.

[0028] S5: Transmits robot driving information. As can be seen, the adjusted local plan controls the robot.

[0029] According to this embodiment, different obstacle stopping distances and obstacle avoidance distances can be implemented according to the different needs of different mobile robots, and the equipment can be effectively prevented from falling into a local optimum solution without increasing the computational burden.

[0030] As described in the background art, when local path planning is performed using the conventional dynamic window method, if an obstacle is present in front of the robot, the desired trajectory must not only move toward the target, but also move away from the obstacle. This contradiction causes the robot to fall into a local minimum. To solve this problem, the following specific improvements are made to the realization of a local path planning method for a mobile robot using adaptive parameters. Sampling is performed in velocity space to obtain multiple sets of velocities. Trajectory prediction is performed for each of the speed sets to obtain a predicted trajectory. Trajectory prediction can be achieved by conventional techniques and will not be described in detail here. The predicted trajectory is evaluated using a preset trajectory evaluation model, and the parameters in the trajectory evaluation model are adjusted to plan the route and obtain the optimal route. The trajectory evaluation model includes a direction cost, a speed difference cost, and a total trajectory cost, and traverses all predicted trajectories corresponding to multiple speed sets to determine the number of predicted trajectories containing obstacles among all predicted trajectories. If the number of predicted trajectories containing obstacles is greater than a set threshold, the score proportion of the direction cost and the score proportion of the speed difference cost in the trajectory evaluation model are reduced, and a sampling speed is calculated based on the updated proportions.

[0031] When using the trajectory evaluation model to evaluate predicted trajectories and plan a path, all predicted trajectories corresponding to multiple speed sets are traversed to determine the number of predicted trajectories containing obstacles among all predicted trajectories, and if the number of predicted trajectories containing obstacles is greater than a set threshold, the proportion of directional cost and the proportion of speed difference cost in the trajectory evaluation model are reduced, and the sampling speed is calculated based on the updated proportions. By adaptively adjusting the proportion of directional cost and the proportion of speed difference cost in the trajectory evaluation model, the problem of the robot falling into a local optimum can be effectively avoided without increasing the computational burden.

[0032] The directional cost A is the angle between the robot's current position and the local target position, and the robot's current target direction, indicating the robot's planned direction. The default minimum value of the score ratio As corresponding to the directional cost is 0.1, and the default value of the directional cost adaptation coefficient Aa is 1. The speed difference cost B is the difference between the robot's current speed v and its maximum speed Mv. The faster the robot is, the closer it is to the target. The default minimum value of the score ratio Bs corresponding to the speed difference cost is 0.1, and the default value of the speed difference cost adaptation coefficient Ba is 1. The total trajectory cost value C is the sum of the corresponding costs in the cost map for all trajectory points at a set of sampling speeds, and the score ratio corresponding to the total trajectory cost value is Cs. Here, the cost map refers to a simplified map that combines the robot's configuration space based on obstacle information in a static map so that the robot can be displayed as a single point on the cost map. It can also be understood as a single obstacle point, and obstacles are expanded based on the robot's shape.

[0033] Here, the value of the total trajectory cost C = 1 - adaptation coefficient Aa for directional cost * score ratio As corresponding to directional cost - adaptation coefficient Ba for speed difference cost * score ratio Bs corresponding to speed difference cost. The score ratio As corresponding to directional cost + score ratio Bs corresponding to speed difference cost + score ratio Cs corresponding to the value of the total trajectory cost = 1. * indicates a multiplication symbol.

[0034] The trajectory evaluation model is the sum of the triple product of the direction cost adaptation coefficient, the direction cost, and the score rate corresponding to the direction cost, plus the triple product of the speed difference cost adaptation coefficient, the speed difference cost, and the score rate corresponding to the speed difference cost, plus the triple product of the total trajectory cost and the score rate corresponding to the total trajectory cost. Specifically, the final trajectory score = direction cost adaptation coefficient Aa * direction cost A * score rate As corresponding to the direction cost + speed difference cost adaptation coefficient Ba * speed difference cost B * score rate Bs corresponding to the speed difference cost + total trajectory cost C * score rate Cs corresponding to the total trajectory cost, where * indicates a multiplication symbol.

[0035] The velocity sampling space is Vs∩Vd, and trajectory prediction is performed for each set of velocities in the sampling space.

number

[0036] In the formula, Vs is the velocity that the robot can achieve in the motion model, sv is the sampling velocity, sw is the sampling angular velocity, Nv is the minimum velocity, Mv is the maximum velocity, sw is the sampling angular velocity, Nw is the minimum angular velocity, Mw is the maximum angular velocity, Vd is the sampling velocity that the equipment can achieve in the robot dynamics model, Cv is the current velocity, Av is the linear acceleration, St is the sampling time, the sampling time St is greater than the control time Ct, Cw is the current angular velocity, and Aw is the angular acceleration.

[0037] For the trajectory points of the first control period in the predicted trajectory, determine whether a collision will occur during emergency stopping, and if a collision occurs, abandon the velocity sampling for that set; otherwise, accumulate the corresponding costs in the cost map for the trajectory points in each control period, and record whether an obstacle exists on the trajectory.

[0038] All sets of speed trajectories are traversed, the total number of which is SUM, and a set of trajectories containing obstacles is determined. The number of trajectories containing obstacles is N. If the size of the set exceeds a threshold, it is determined that the device is entering an obstacle-dense area, and the speed difference and direction ratio are reduced. Using a trajectory evaluation model, the trajectory scores are normalized, the corresponding score ratios are updated, and the optimal sampling speed is calculated. As can be seen, each sampled speed and angular velocity can generate one trajectory information, and each trajectory information can receive one score. The sampling speed and angular velocity corresponding to the trajectory with the smallest score are the optimal speed. The adaptation coefficients for the direction cost and the speed difference cost are both set to the maximum value between 0.1 and the predicted value, where the predicted value is the ratio of the difference between the total number of predicted trajectories and the number of trajectories containing obstacles to the total number of predicted trajectories. Specifically, the adaptive coefficient of the direction cost is Aa=max((SUM-N) / SUM, 0.1), and the adaptive coefficient of the speed difference cost is Ba=max((SUM-N) / SUM, 0.1).

[0039] If the size of the set does not exceed the threshold, the current point and the local target point are connected. If there is obstacle information on the connecting line, the route is re-planned by the global planner, limiting the route search time of the global planner to prevent the route search time from being too long. The route is put into the local planned target point set, the local target point is updated, and substituted into the trajectory evaluation model, and the optimal sampling rate is calculated.

[0040] The optimal sampling rate is transmitted to the chassis of the mobile robot to realize control and path planning for the robot.

[0041] Example 2 In this embodiment, a sampling module configured to sample in velocity space to obtain a plurality of velocity sets; a prediction module configured to perform trajectory prediction for each of the speed sets to obtain a predicted trajectory; a planning module configured to evaluate a predicted trajectory according to a preset trajectory evaluation model, and to plan a route by adjusting parameters in the trajectory evaluation model to obtain an optimal route; The trajectory evaluation model includes a direction cost, a speed difference cost, and a total trajectory cost, and all predicted trajectories corresponding to a plurality of speed sets are traversed to determine the number of predicted trajectories containing obstacles among all the predicted trajectories, and if the number of predicted trajectories containing obstacles is greater than a set threshold, the score proportions of the direction cost and the speed difference cost in the trajectory evaluation model are reduced, and a sampling speed is calculated based on the updated proportions.

[0042] The operation method of the system is the same as the local path planning method for a mobile robot using adaptive parameters in the first embodiment, and will not be described in detail here.

[0043] Example 3 This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method for local path planning for a mobile robot with adaptive parameters described in embodiment 1.

[0044] Example 4 This embodiment provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, which, when executed by the processor, realizes the steps of the local path planning method for a mobile robot using adaptive parameters described in embodiment 1.

[0045] The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention. Those skilled in the art may make various modifications and changes to the present invention. Any modifications, equivalent replacements, improvements, etc. made without departing from the spirit and principles of the present invention shall all be included in the scope of protection of the present invention.

Claims

1. sampling in velocity space to obtain a plurality of velocity sets; performing trajectory prediction for each of the speed sets to obtain a predicted trajectory; evaluating the predicted trajectory using a preset trajectory evaluation model, and adjusting parameters in the trajectory evaluation model to plan a route and obtain an optimal route; The trajectory evaluation model includes a direction cost, a speed difference cost, and a total trajectory cost; all predicted trajectories corresponding to a plurality of speed sets are traversed to determine the number of predicted trajectories containing obstacles among all the predicted trajectories; if the number of predicted trajectories containing obstacles is greater than a set threshold, the score proportions of the direction cost and the speed difference cost in the trajectory evaluation model are reduced; and the sampling speed is calculated based on the updated proportions.

2. 2. The local path planning method for a mobile robot using adaptive parameters according to claim 1, wherein the direction cost is the angle between the current position of the robot and the local target position and the current target direction of the robot, the speed difference cost is the difference between the current speed of the robot and the maximum speed of the robot, and the total trajectory cost is the sum of corresponding costs in the cost map of all trajectory points at a set of sampling speeds.

3. 2. The local path planning method for a mobile robot using adaptive parameters according to claim 1, wherein the trajectory evaluation model is a sum of a product of an adaptation coefficient for a direction cost, a direction cost, and a score rate corresponding to the direction cost, plus a product of an adaptation coefficient for a speed difference cost, a speed difference cost, and a score rate corresponding to the speed difference cost, plus a product of a total trajectory cost value and a score rate corresponding to the total trajectory cost value.

4. 4. The local path planning method for a mobile robot using adaptive parameters according to claim 3, wherein the adaptive coefficients of the direction cost and the speed difference cost each take the maximum value between 0.1 and a predicted value, and the predicted value is the ratio of the difference between the total number of predicted trajectories and the number of trajectories containing obstacles to the total number of predicted trajectories.

5. 2. The local path planning method for a mobile robot using adaptive parameters according to claim 1, characterized in that: it is determined whether a collision occurs when an emergency stop is performed on a trajectory point in the first control cycle on the predicted trajectory; if so, the corresponding velocity sampling is abandoned; if not, the corresponding costs in the cost map of the trajectory points in each control cycle are accumulated.

6. 2. The local path planning method for a mobile robot using adaptive parameters according to claim 1, characterized in that if the number of predicted trajectories containing obstacles is less than a set threshold, the current point of the robot and the local target point are connected; if obstacle information is present on the line connecting the current point of the robot and the local target point, the path is re-planned using a global planner, the planned path is entered into a local planning target point set, the local target point is updated, and substituted into a trajectory evaluation model, and an optimal sampling rate is calculated.

7. Use the global planner to find route points to your destination, The Floyd path smoothing algorithm sets local path planning key points, Select the target point from the key points, 7. The local path planning method for a mobile robot using adaptive parameters according to claim 6, characterized in that, when the equipment is traveling, a local map and a trajectory evaluation model are constructed, and the local plan is adjusted based on whether an obstacle exists between the current point and the target point.

8. a sampling module configured to sample in velocity space to obtain a plurality of velocity sets; a prediction module configured to perform trajectory prediction for each of the speed sets to obtain a predicted trajectory; a planning module configured to evaluate a predicted trajectory according to a preset trajectory evaluation model, and to plan a route by adjusting parameters in the trajectory evaluation model to obtain an optimal route; The trajectory evaluation model includes a direction cost, a speed difference cost, and a total trajectory cost, and all predicted trajectories corresponding to a plurality of speed sets are traversed to determine the number of predicted trajectories containing obstacles among all the predicted trajectories, and if the number of predicted trajectories containing obstacles is greater than a set threshold, the score proportions of the direction cost and the speed difference cost in the trajectory evaluation model are reduced, and the sampling speed is calculated based on the updated proportions.

9. A computer-readable storage medium having stored thereon a computer program that, when executed by a processor, implements the steps of the method for local path planning for a mobile robot using adaptive parameters according to any one of claims 1 to 7.

10. 8. An electronic device comprising: a memory; a processor; and a computer program stored in the memory and executable on the processor, the program implementing, when executed by the processor, the steps of the method for local path planning for a mobile robot using adaptive parameters according to any one of claims 1 to 7.

Images