A robotic arm path planning method and system based on adaptive adjustment of environmental parameters

Abstract

The present invention relates to the field of path planning technology, and in particular to a method and system for robotic arm path planning based on adaptive adjustment of environmental parameters. The method comprises: constructing a random tree with the robotic arm end position as the starting node and the target object position as the target node, setting a search radius to dynamically acquire obstacle data; dynamically adjusting the step size based on the obstacle data, introducing an adaptive sampling strategy, and generating an initial path by selecting the optimal parent node and reconnecting surrounding nodes; smoothing the initial path, adjusting the offset point, optimizing the target node search area, and constraining the path direction angle to obtain the final path; converting the path node coordinates into the robotic arm's physical coordinate system coordinates, and driving the robotic arm to perform obstacle avoidance and grasping tasks. The present invention significantly improves path planning efficiency, shortens planning time, optimizes path quality, and enhances safety and stability in complex environments.

Description

Technical Field

[0001] The present invention relates to the technical field of path planning, and in particular to a method and system for robot arm path planning based on adaptive adjustment of environmental parameters. Background Art

[0002] In key technical fields such as robotic arm motion control, autonomous driving, and logistics and transportation, path planning is the core technology to ensure the efficient operation of the system. Its performance directly determines the operating efficiency and reliability of each system. At present, traditional path planning algorithms are still widely used in path planning scenarios. Taking the A* algorithm as an example, it can effectively search for the global optimal path in a simple static environment, but when faced with complex obstacle distribution or dynamically changing environments, the algorithm needs to traverse a large number of nodes, causing the amount of calculation to increase exponentially, making it difficult to meet the needs of real-time path planning. The Dijkstra algorithm also faces an efficiency bottleneck. Its high time complexity significantly increases the planning time when processing large-scale map data, making it unsuitable for application scenarios with extremely high real-time requirements.

[0003] To address the shortcomings of traditional algorithms, the existing improved Informed RRT* algorithm, based on the RRT algorithm, introduces an elliptical sampling strategy to optimize the search range, which improves search efficiency to a certain extent. However, this algorithm still has significant technical flaws: First, it lacks adaptability to dynamic environments and cannot quickly respond to environmental changes and replan the path; second, it lacks an effective path safety assurance mechanism, making it difficult to avoid potential collision risks; and third, its robustness in the endpoint area is poor, making it prone to path oscillation or failure to accurately converge to the target point. These issues limit the application and development of existing path planning algorithms in complex dynamic scenarios. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides a robot arm path planning method and system based on adaptive adjustment of environmental parameters, the method comprising the following steps:

[0005] S1: Taking the current position of the end of the robot arm as the starting node, the position of the target object to be grasped as the target node, and the starting node as the starting point, a random tree containing path nodes between the starting node and the target node is constructed. In the random tree, with each current node as the center, a search radius is set to dynamically obtain obstacle data within the range;

[0006] S2: Dynamically adjust the step size of the end of the robotic arm based on the obstacle data. At the same time, introduce an adaptive sampling strategy. When selecting the parent node of a new node, prioritize nearby nodes with the lowest total cost and no collisions. Reconnect the surrounding nodes to generate an initial path.

[0007] S3: Smoothing the initial path, reducing turning points in the initial path, dynamically adjusting path offset points, prioritizing offset points that are close to the target node and have no collisions, and obtaining the final path by optimizing the search area of ​​the target node and constraining the angle between the path direction and the target node direction;

[0008] S4: Based on the final path, path node coordinates are generated, the path node coordinates are converted into path coordinates in the physical coordinate system of the robotic arm, and based on the path coordinates in the physical coordinate system of the robotic arm, the robotic arm is driven to perform the obstacle avoidance and grasping task.

[0009] In one embodiment of the present invention, in S2, the method for generating the initial path is as follows:

[0010] S21: Calculate the straight path length between the starting node and the target node, calculate the target sampling probability according to the straight path length and the current path length, and generate sampling points according to the target sampling probability;

[0011] S22: Find the first node closest to the sampling point in the random tree , from the first node Extend the step length to the sampling point to generate the second node ;

[0012] S23: traverse the first node and the second node Are there any obstacles on the path between:

[0013] If so, return to S21 to resample;

[0014] If not, proceed to step S24;

[0015] S24: traverse the second node Get the neighbor node set in the step area of ​​an integer multiple of , for the neighbor node set Each node in , calculate each node To the second node Total cost :

[0016] ,

[0017] in, Representation node The length of the path to the starting node, Represents the second node and nodes The length of the path between

[0018] S25: Select the total cost The smallest node with no collision path is the parent node of the second node , the second node Add the random tree and update the path cost of the second node to:

[0019] ,

[0020] in, Representation node The length of the path to the starting node, Represents the second node and nodes The length of the path between

[0021] S26: traverse the second node All nodes within a radius r , calculation through each node The second node Candidate cost ,like And the path has no collision, update the node The parent node is , and update the node The node cost is ;

[0022] S27: Repeat steps S21 to S26 to obtain an initial path.

[0023] In one embodiment of the present invention, in S21, the target sampling probability is calculated according to the straight path length and the current path length, and the method for generating sampling points according to the target sampling probability is as follows:

[0024] Setting global variables Represents the current optimal path length, and its initial value is set to , define the path length calculation function , and obtain the total length by accumulating the Euclidean distance of each path segment:

[0025] ,

[0026] in, represents the path node coordinates, N represents the number of nodes, represents the Euclidean distance;

[0027] When the initial target sampling probability Sampling target points to generate new candidate paths When the path length calculation function is used to calculate the new candidate path Length ,like Less than , then use renew The value of and synchronously update the target sampling probability :

[0028] ,

[0029] in, Indicates the starting node and target node The straight-line distance between

[0030] Generate a random number between 0 and 1 ,like , then the target node is used as the sampling point; otherwise, in the elliptical sampling area Random sampling is performed within to obtain sampling points; in each iterative calculation, if it is found that the current path length is less than the one after the last iterative update, , update the elliptical sampling area using the current shortest path Parameters.

[0031] In one embodiment of the present invention, the candidate cost :

[0032] ,

[0033] in, Representation node The length of the path to the starting node, Represents the second node and nodes The length of the path between .

[0034] In one embodiment of the present invention, the obstacle data includes obstacle type, obstacle information, and obstacle quantity, and the acquisition method thereof is as follows:

[0035] When the obstacle is a circular obstacle, the obstacle information also includes the coordinates and radius of the center point of the circular obstacle. Within the search radius, if the distance from the current node to the center point of the circular obstacle is satisfy When the condition is met, the number of obstacles is accumulated by 1; represents the search radius, Indicates the radius of the circular obstacle;

[0036] When the obstacle is a polygonal obstacle, the obstacle information includes the coordinates of each vertex of the polygonal obstacle surface, the distance from the current node to any vertex of the polygonal obstacle within the search radius, satisfy When the condition is met, the number of obstacles is accumulated by 1; Represents the search radius.

[0037] In one embodiment of the present invention, the obstacle data also includes obstacle density , which is calculated as follows:

[0038] ,

[0039] in, Indicates the total number of obstacles.

[0040] In one embodiment of the present invention, a method for dynamically adjusting the step length of the end of the robotic arm according to the obstacle data is as follows:

[0041] According to the obstacle density , dynamically adjust the step length of the end of the robotic arm , and its calculation formula is:

[0042] ,

[0043] in, Indicates the minimum step size in the obstacle-dense area. Indicates the maximum step size in the obstacle sparse area.

[0044] In one embodiment of the present invention, in S3, the method for obtaining the final path is as follows:

[0045] Using the nodes of the initial path as interpolation points, setting the smoothing factor of the curve to control the curvature of the curve, interpolating the line segments between adjacent nodes to generate a continuous curve, replacing the original discrete node path, and generating a smoothed initial path;

[0046] Each node in the smoothed initial path As the center, polar coordinate parameterization is used to generate a set of candidate offset points :

[0047] Set the initial offset and offset angle intervals , that is, every Generate 1 direction, generate the number of directions K, and the corresponding angle set ;

[0048] For each direction , the coordinates of the offset point are: , forming a set of candidate offset points ;

[0049] For the candidate offset point set , prioritize the distance to the target node Shortest offset point: , Indicates offset point and target node The Euclidean distance between

[0050] At the same time, the offset point Satisfy the constraints, including space occupancy legality constraints and line segment collision detection constraints, where the space occupancy legality constraints are: offset point Meeting barrier-free occupancy requirements: ,in represents the space occupied by the i-th obstacle, Represents the total space occupied by m obstacles;

[0051] The line segment collision detection constraint condition is: define the second to last node of the path as , ensure that the node and nodes The connecting line between No intersection with the obstacle set: , m represents the total number of obstacles;

[0052] If the current offset point does not meet the constraint conditions, the offset is increased according to the set geometric progression to generate a new offset point set, and the search area radius for determining the target node is adaptively increased. ;

[0053] When any node enters the search area of ​​the target node, it satisfies the angle between the path direction from the previous node to the node and the direction from the node to the target node. If the angle is less than the preset threshold, the final path is obtained.

[0054] In one embodiment of the present invention, the radius of the search area for determining the target node is adaptively increased. The method is as follows:

[0055] ,

[0056] in, Indicates the radius of the initial target node area, max_iter indicates the maximum number of iterations, Indicates the current iteration number.

[0057] In one embodiment of the present invention, the angle The calculation method is as follows:

[0058] ,

[0059] in, Indicates the path direction vector from the previous node of the current node to this node. Indicates the direction vector from the current node to the target node.

[0060] Based on the same inventive concept, the present invention also provides a robotic arm path planning system based on adaptive adjustment of environmental parameters, the robotic arm path planning system based on adaptive adjustment of environmental parameters includes the following modules:

[0061] An initialization module is used to construct a random tree containing path nodes between the starting node and the target node, with the current position of the robot arm end as the starting node and the position of the target object to be grasped as the target node. In the random tree, a search radius is set with each current node as the center, and obstacle data within the range is dynamically obtained;

[0062] A path generation module is used to dynamically adjust the step size of the end of the robotic arm based on the obstacle data. At the same time, an adaptive sampling strategy is introduced to select the parent node of a new node, giving priority to the nearby node with the minimum total cost and no collision, and reconnecting the surrounding nodes to generate an initial path;

[0063] a path optimization module for smoothing the initial path, reducing turning points in the initial path, dynamically adjusting path offset points, giving priority to offset points that are close to the target node and have no collisions, and obtaining the final path by optimizing the search area of ​​the target node and constraining the angle between the path direction and the target node direction;

[0064] The task execution module is used to generate path node coordinates based on the final path, convert the path node coordinates into path coordinates in the physical coordinate system of the robotic arm, and drive the robotic arm to perform the obstacle avoidance and grasping task based on the path coordinates in the physical coordinate system of the robotic arm.

[0065] The present invention also provides a computer storage medium, which stores a computer software product. The computer software product includes several instructions for enabling a computer device to execute the robotic arm path planning method based on adaptive adjustment of environmental parameters.

[0066] The above technical solution of the present invention has the following advantages over the prior art:

[0067] The present invention uses a dynamic step-size adjustment mechanism to flexibly change the step size according to the density of obstacles, ensuring safety through detailed exploration in dense areas and improving efficiency through large-scale expansion in sparse areas. It introduces an adaptive sampling strategy to dynamically adjust the sampling probability of the target point according to the current optimal path length, guiding the search tree to quickly grow toward the target and accelerate convergence. It optimizes the node connection and reconnection mechanism, selects the parent node with the lowest total cost and no collisions, and reconnects the surrounding nodes to reduce the path cost. It smoothes and offsets the preliminary path to reduce turning points, giving priority to offset points near the end point and no collisions to improve the path quality. It also improves the success rate of the end point connection by dynamically expanding the radius of the end point area and combining it with directional alignment constraints. The overall solution significantly improves the efficiency of path planning, shortens planning time, optimizes path quality, and enhances safety and stability in complex environments. It has the advantages of high efficiency, precision, safety, and strong portability, providing strong support for industrial production automation. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below based on specific embodiments of the present invention in conjunction with the accompanying drawings, wherein:

[0069] Figure 1 A schematic flow chart of a method for robotic arm path planning based on adaptive adjustment of environmental parameters is provided in an embodiment of the present invention;

[0070] Figure 2 1 is a schematic diagram of the structure of a robot arm path planning system based on adaptive adjustment of environmental parameters provided in an embodiment of the present invention;

[0071] Description of the accompanying drawings in the specification: 100, initialization module; 200, path generation module; 300, path optimization module; 400, task execution module. DETAILED DESCRIPTION

[0072] The present invention will be further described below with reference to the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it. However, the embodiments are not intended to limit the present invention.

[0073] Example 1:

[0074] like Figure 1 As shown, the present invention provides a robotic arm path planning method based on adaptive adjustment of environmental parameters. Based on an improved RRT* algorithm, the method improves the efficiency of path planning, reduces planning time, optimizes path quality, enhances the safety and stability of path planning in complex obstacle environments, and meets the requirements for efficient and accurate path planning in different scenarios. Specifically, the method includes the following steps:

[0075] S1: Taking the current position of the end of the robot arm as the starting node, the position of the target object to be grasped as the target node, and the starting node as the starting point, a random tree containing path nodes between the starting node and the target node is constructed. In the random tree, with each current node as the center, a search radius is set to dynamically obtain obstacle data within the range;

[0076] S2: Dynamically adjust the step size of the end of the robotic arm based on the obstacle data. At the same time, introduce an adaptive sampling strategy. When selecting the parent node of a new node, prioritize nearby nodes with the lowest total cost and no collisions. Reconnect the surrounding nodes to generate an initial path.

[0077] S3: Smoothing the initial path, reducing turning points in the initial path, dynamically adjusting path offset points, prioritizing offset points that are close to the target node and have no collisions, and obtaining the final path by optimizing the search area of ​​the target node and constraining the angle between the path direction and the target node direction;

[0078] S4: Based on the final path, path node coordinates are generated, the path node coordinates are converted into path coordinates in the physical coordinate system of the robotic arm, and based on the path coordinates in the physical coordinate system of the robotic arm, the robotic arm is driven to perform the obstacle avoidance and grasping task.

[0079] Furthermore, in S1, the current posture of the end effector of the robot arm is used as the starting node , the spatial coordinates of the target object to be grasped are defined as , the random tree is initialized with only a single root node ,and The target state of path planning is initialized independently and is not included in the initial node set. Build a random tree containing the path nodes between the starting node and the target node for the root node, and set the search radius with each current node as the center in the tree structure , dynamically obtain obstacle data within the range, where the obstacle data includes obstacle type, obstacle information, obstacle quantity and obstacle density. The acquisition method is as follows:

[0080] When the obstacle is a circular obstacle, the obstacle information also includes the center point coordinates and radius of the circular obstacle. If the distance between the current node and the center point of the circular obstacle is satisfy When the condition is met, the number of obstacles is accumulated by 1; Indicates the radius of the circular obstacle;

[0081] When the obstacle is a polygonal obstacle, the obstacle information includes the coordinates of each vertex of the polygonal obstacle surface, the distance from the current node to any vertex of the polygonal obstacle within the search radius, satisfy When the condition is met, the number of obstacles is accumulated by 1.

[0082] Specifically, the total number of obstacles obtained by statistics , calculate the obstacle density , the formula is as follows:

[0083] .

[0084] Furthermore, in this embodiment, in S2, to improve the efficiency and reliability of path generation in complex environments, the method for generating the initial path is as follows:

[0085] S21: Calculating the straight path length between the starting node and the target node, calculating the target sampling probability according to the straight path length and the current path length, and generating sampling points according to the target sampling probability, including:

[0086] Setting global variables Represents the current optimal path length, and its initial value is set to , define the path length calculation function , and obtain the total length by accumulating the Euclidean distance of each path segment:

[0087] ,

[0088] in, represents the path node coordinates, N represents the number of nodes, represents the Euclidean distance;

[0089] When the initial target sampling probability Sampling target points to generate new candidate paths When the path length calculation function is used to calculate the new candidate path Length ,like Less than , then use renew The value of and synchronously update the target sampling probability :

[0090] ;

[0091] Generate a random number between 0 and 1 ,like , then the target node is used as the sampling point; otherwise, in the elliptical sampling area Random sampling is performed within to obtain sampling points; in each iterative calculation, if it is found that the current path length is less than the one after the last iterative update, , update the elliptical sampling area using the current shortest path The specific steps are as follows:

[0092] In the first iteration, the elliptical sampling area Initial path length As the major axis 2a, the focal length 2c is determined by the Euclidean distance between the start and end points:

[0093]

[0094]

[0095] At this time, the elliptical sampling area Defined as: , X is the sampling point;

[0096] If each iteration finds that the current path length is less than the updated path length in the previous iteration, , then update the elliptical sampling area according to the following rules Parameters:

[0097]

[0098] At this time, the elliptical sampling area shrinks to:

[0099] , k represents the number of iterations.

[0100] S22: Find the first node closest to the sampling point in the random tree , from the first node Extend the step length to the sampling point to generate the second node ,The step size is dynamically adjusted according to the obstacle density. When the obstacles are dense, the step size is reduced for fine exploration, and when the obstacles are sparse, the step size is increased to accelerate expansion;

[0101] Specifically, according to the obstacle density , dynamically adjust the step length of the end of the robotic arm , and its calculation formula is:

[0102] ,

[0103] in, Indicates the minimum step size in the obstacle-dense area. Indicates the maximum step size in the obstacle sparse area;

[0104] S23: traverse the first node and the second node Are there any obstacles on the path between:

[0105] If so, return to S21 to resample;

[0106] If not, proceed to step S24;

[0107] S24: Using the second node As the center, in the integer multiple step size area (such as the radius is Get the neighbor node set within the circular area) For the neighbor node set Each node in , calculate each node To the second node Total cost :

[0108] ,

[0109] in, Representation node The length of the path to the starting node, Represents the second node and nodes The length of the path between

[0110] S25: From the neighbor node set , select the total cost The smallest node with no collision path is the parent node of the second node , the second node Add to the random tree and update the path cost of the second node to:

[0111] ,

[0112] in, Representation node The length of the path to the starting node, Represents the second node and nodes The length of the path between

[0113] S26: traverse the second node All nodes within a radius r , calculation through each node The second node Candidate cost ,like And the path has no collision, update the node The parent node is , and synchronously update the nodes The node cost is ; Wherein, the candidate cost The calculation method is as follows:

[0114] ,

[0115] in, Representation node The length of the path to the starting node, Represents the second node and nodes The length of the path between

[0116] S27: Repeat steps S21 to S26 until the termination condition is met (such as finding a feasible node in the neighborhood of the target node or reaching the maximum number of iterations), and finally generate an initial path containing node coordinates and path costs.

[0117] Furthermore, in this embodiment, in S3, the method for obtaining the final path is as follows:

[0118] The node set of the initial path is the interpolation control point, sampling band tension parameter The cubic spline interpolation algorithm is used for path smoothing:

[0119] For adjacent nodes and , construct a parametric spline curve , ,satisfy: , ;

[0120] The tangent vector is given by the tension parameter control: ,

[0121] ,in , When degenerates into a natural spline, Enhance tangent continuity when

[0122] Generate a smooth curve point set by equidistant sampling, replace the original discrete node path, reduce the number of path turning points from n-1 to a theoretically continuous and differentiable smooth curve, and generate a smoothed initial path;

[0123] Each node in the smoothed initial path As the center, polar coordinate parameterization is used to generate a set of candidate offset points :

[0124] Set the initial offset Unit, offset angle interval = 45°), that is, one direction is generated every 45°, the number of generated directions K=8, and the corresponding angle set .

[0125] For each direction , the coordinates of the offset point are: , forming a set of candidate offset points ;

[0126] A goal-oriented heuristic search strategy is used to select candidate offset point sets. Do the following:

[0127] Prioritize the distance to the target node Shortest offset point: , Indicates offset point and target node The Euclidean distance between

[0128] At the same time, the offset point Satisfy the constraints, including space occupancy legality constraints and line segment collision detection constraints, where the space occupancy legality constraints are: offset point Meeting barrier-free occupancy requirements: ,in represents the space occupied by the i-th obstacle, Represents the total space occupied by m obstacles;

[0129] The line segment collision detection constraint condition is: define the second to last node of the path as , ensure that the node and nodes The connecting line between No intersection with the obstacle set: , m represents the total number of obstacles.

[0130] If the current offset point does not meet the constraints, the offset is calculated in geometric progression. Incrementally, generate a new offset point set, is the number of increments, Indicates the initial offset, Indicates the offset after the kth increment;

[0131] At the same time, the radius of the search area for entering the target node is adaptively increased :

[0132] ,

[0133] in, Indicates the radius of the initial target node area, max_iter indicates the maximum number of iterations, Represents the current iteration number; this formula ensures that the search domain shrinks exponentially as the iteration advances, and focuses on the target point at the end.

[0134] When any node enters the target node The search area of ​​​​the current node is defined The path direction vector , target direction vector , For the current node The previous node of the node, calculate the path direction from the previous node of the node to the node The node points to the target node. The angle between :

[0135] ,

[0136] in, Indicates the path direction vector from the previous node of the current node to this node. Indicates the direction vector of the current node pointing to the target node;

[0137] Set the preset angle threshold In this embodiment ,when When , the current path is confirmed to be the final path; when When , insert an intermediate node at the end of the path , is the step size coefficient, and adjusts the target node The search area is an elliptical search area:

[0138] , where the long axis , short axis , is the scaling factor to guide the path to converge toward the target node.

[0139] Through the above-mentioned smoothing processing, offset adjustment and direction constraints, a final path is generated that is collision-free, smooth and accurately aligned with the target, meeting the engineering application requirements of the robot arm's obstacle avoidance and grasping.

[0140] Example 2:

[0141] like Figure 2As shown, based on the same inventive concept as that of the first embodiment, the present invention further provides a robot arm path planning system based on adaptive adjustment of environmental parameters, which specifically includes the following modules: an initialization module 100, a path generation module 200, a path optimization module 300 and a task execution module 400;

[0142] The initialization module 100 is configured to construct a random tree containing path nodes between the starting node and the target node, with the current position of the end of the robotic arm as the starting node and the position of the target object to be grasped as the target node. In the random tree, a search radius is set with each current node as the center, and obstacle data within the range is dynamically acquired.

[0143] The path generation module 200 is used to dynamically adjust the step size of the end of the robotic arm based on the obstacle data, and introduce an adaptive sampling strategy to prioritize nearby nodes with the lowest total cost and no collision when selecting the parent node of a new node, and reconnect the surrounding nodes to generate an initial path;

[0144] The path optimization module 300 is used to smooth the initial path, reduce turning points in the initial path, dynamically adjust path offset points, give priority to offset points that are close to the target node and have no collisions, and obtain the final path by optimizing the search area of ​​the target node and constraining the angle between the path direction and the target node direction;

[0145] The task execution module 400 is used to generate path node coordinates based on the final path, convert the path node coordinates into path coordinates in the physical coordinate system of the robotic arm, and drive the robotic arm to perform the obstacle avoidance and grasping task based on the path coordinates in the physical coordinate system of the robotic arm.

[0146] The present embodiment proposes a robotic arm path planning system based on adaptive adjustment of environmental parameters, which is used to implement the aforementioned robotic arm path planning method based on adaptive adjustment of environmental parameters. Therefore, the specific implementation method of the robotic arm path planning system based on adaptive adjustment of environmental parameters can be seen in the aforementioned embodiment part of the robotic arm path planning method based on adaptive adjustment of environmental parameters. For example, the initialization module 100, the path generation module 200, the path optimization module 300 and the task execution module 400 are respectively used to implement steps S1, S2, S3 and S4 in the robotic arm path planning method based on adaptive adjustment of environmental parameters in embodiment one. Therefore, its specific implementation method can refer to the description of the corresponding embodiments of each part. In order to avoid redundancy, it will not be repeated here.

[0147] Example 3:

[0148] The present invention also provides a computer storage medium, which stores a computer software product. The computer software product includes several instructions for enabling a computer device to execute the robotic arm path planning method based on adaptive adjustment of environmental parameters described in Example 1.

[0149] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0150] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0151] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0152] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0153] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.

Claims

1. A robot arm path planning method based on adaptive adjustment of environmental parameters, characterized in that: The following steps are involved: S1: With the current position of the robot arm end as the starting node and the position of the target object to be grasped as the target node, a random tree containing path nodes between the starting node and the target node is constructed with the starting node as the starting point. In the random tree, with each current node as the center, a search radius is set to dynamically obtain obstacle data within the range; wherein, the obstacle data includes obstacle type, obstacle information, obstacle number and obstacle density, and the acquisition method is as follows: When the obstacle is a circular obstacle, the obstacle information also includes the coordinates and radius of the center point of the circular obstacle. Within the search radius, if the distance from the current node to the center point of the circular obstacle is satisfy When the condition is met, the number of obstacles is accumulated by 1; represents the search radius, Indicates the radius of the circular obstacle; When the obstacle is a polygonal obstacle, the obstacle information includes the coordinates of each vertex of the polygonal obstacle surface, the distance from the current node to any vertex of the polygonal obstacle within the search radius, satisfy When the condition is met, the number of obstacles is accumulated by 1; represents the search radius; The obstacle density The calculation method is: , Indicates the total number of obstacles; S2: Dynamically adjust the step size of the end of the robotic arm based on the obstacle data. At the same time, introduce an adaptive sampling strategy. When selecting the parent node of a new node, prioritize nearby nodes with the lowest total cost and no collisions. Reconnect the surrounding nodes to generate an initial path. The method for dynamically adjusting the step size of the end of the robotic arm based on the obstacle data is as follows: According to the obstacle density , dynamically adjust the step length of the end of the robotic arm , and its calculation formula is: ， in, Indicates the minimum step size in the obstacle-dense area. Indicates the maximum step size in the obstacle sparse area; S3: Smoothing the initial path, reducing turning points in the initial path, dynamically adjusting path offset points, prioritizing offset points that are close to the target node and have no collisions, and obtaining the final path by optimizing the search area of ​​the target node and constraining the angle between the path direction and the target node direction; S4: Based on the final path, path node coordinates are generated, the path node coordinates are converted into path coordinates in the physical coordinate system of the robotic arm, and based on the path coordinates in the physical coordinate system of the robotic arm, the robotic arm is driven to perform the obstacle avoidance and grasping task.

2. The robot arm path planning method based on adaptive adjustment of environmental parameters according to claim 1, characterized in that: In S2, the method for generating the initial path is as follows: S21: Calculate the straight path length between the starting node and the target node, calculate the target sampling probability according to the straight path length and the current path length, and generate sampling points according to the target sampling probability; S22: Find the first node closest to the sampling point in the random tree , from the first node Extend the step length to the sampling point to generate the second node ; S23: traverse the first node and the second node Are there any obstacles on the path between: If so, return to S21 to resample; If not, proceed to step S24; S24: traverse the second node Get the neighbor node set in the step area of ​​an integer multiple of , for the neighbor node set Each node in , calculate each node To the second node Total cost : ， in, Representation node The length of the path to the starting node, Represents the second node and nodes The length of the path between S25: Select the total cost The smallest node with no collision path is the parent node of the second node , the second node Add to the random tree and update the path cost of the second node to: ， in, Representation node The length of the path to the starting node, Represents the second node and nodes The length of the path between S26: traverse the second node All nodes within a radius r , calculation through each node The second node Candidate cost ,like And the path has no collision, update the node The parent node is , and update the node The node cost is ; S27: Repeat steps S21 to S26 to obtain an initial path.

3. The robot arm path planning method based on adaptive adjustment of environmental parameters according to claim 2, characterized in that: In S21, the target sampling probability is calculated according to the straight path length and the current path length, and the method for generating sampling points according to the target sampling probability is as follows: Setting global variables Represents the current optimal path length, and its initial value is set to , define the path length calculation function , and obtain the total length by accumulating the Euclidean distance of each path segment: ， in, represents the path node coordinates, N represents the number of nodes, represents the Euclidean distance; When the initial target sampling probability Sampling target points to generate new candidate paths When the path length calculation function is used to calculate the new candidate path Length ,like Less than , then use renew The value of and synchronously update the target sampling probability : ， in, Indicates the starting node and target node The straight-line distance between Generate a random number between 0 and 1 ,like , then the target node is used as the sampling point; otherwise, in the elliptical sampling area Random sampling is performed within to obtain sampling points; in each iterative calculation, if it is found that the current path length is less than the one after the last iterative update, , update the elliptical sampling area using the current shortest path Parameters.

4. The robot arm path planning method based on adaptive adjustment of environmental parameters according to claim 2, characterized in that: The candidate cost The calculation method is as follows: ， in, Representation node The length of the path to the starting node, Represents the second node and nodes The length of the path between .

5. The robot arm path planning method based on adaptive adjustment of environmental parameters according to claim 1, characterized in that: In S3, the method to obtain the final path is as follows: Using the nodes of the initial path as interpolation points, setting the smoothing factor of the curve to control the curvature of the curve, interpolating the line segments between adjacent nodes to generate a continuous curve, replacing the original discrete node path, and generating a smoothed initial path; Each node in the smoothed initial path As the center, polar coordinate parameterization is used to generate a set of candidate offset points : Set the initial offset and offset angle intervals , that is, every Generate 1 direction, generate the number of directions K, and the corresponding angle set ; For each direction , the coordinates of the offset point are: , forming a set of candidate offset points ; For the candidate offset point set , prioritize the distance to the target node Shortest offset point: , Indicates offset point and target node The Euclidean distance between At the same time, the offset point Satisfy the constraints, including space occupancy legality constraints and line segment collision detection constraints, where the space occupancy legality constraints are: offset point Meeting barrier-free occupancy requirements: ,in represents the space occupied by the i-th obstacle; Represents the total space occupied by m obstacles; The line segment collision detection constraint condition is: define the second to last node of the path as , ensure that the node and nodes The connecting line between No intersection with the obstacle set: , m represents the total number of obstacles; If the current offset point does not meet the constraint conditions, the offset is increased according to the set geometric progression to generate a new offset point set, and the search area radius for determining the target node is adaptively increased. ; When any node enters the search area of ​​the target node, it satisfies the angle between the path direction from the previous node to the node and the direction from the node to the target node. If the angle is less than the preset threshold, the final path is obtained.

6. The robot arm path planning method based on adaptive adjustment of environmental parameters according to claim 5, characterized in that: Adaptively increase the search area radius for determining the target node The method is as follows: ， in, Indicates the radius of the initial target node area, max_iter indicates the maximum number of iterations, Indicates the current iteration number.

7. The robot arm path planning method based on adaptive adjustment of environmental parameters according to claim 5, characterized in that: The angle The calculation method is as follows: ， in, Indicates the path direction vector from the previous node of the current node to this node. Indicates the direction vector from the current node to the target node.

8. A robotic arm path planning system based on adaptive adjustment of environmental parameters, characterized in that: The steps of implementing the method for robotic arm path planning based on adaptive adjustment of environmental parameters according to any one of claims 1 to 7 are as follows: the robotic arm path planning system based on adaptive adjustment of environmental parameters comprises the following modules: An initialization module is used to construct a random tree containing path nodes between the starting node and the target node, with the current position of the robot arm end as the starting node and the position of the target object to be grasped as the target node. In the random tree, a search radius is set with each current node as the center, and obstacle data within the range is dynamically obtained; A path generation module is used to dynamically adjust the step size of the end of the robotic arm based on the obstacle data. At the same time, an adaptive sampling strategy is introduced to select the parent node of a new node, giving priority to the nearby node with the minimum total cost and no collision, and reconnecting the surrounding nodes to generate an initial path; a path optimization module for smoothing the initial path, reducing turning points in the initial path, dynamically adjusting path offset points, giving priority to offset points that are close to the target node and have no collisions, and obtaining the final path by optimizing the search area of ​​the target node and constraining the angle between the path direction and the target node direction; The task execution module is used to generate path node coordinates based on the final path, convert the path node coordinates into path coordinates in the physical coordinate system of the robotic arm, and drive the robotic arm to perform the obstacle avoidance and grasping task based on the path coordinates in the physical coordinate system of the robotic arm.

9. A computer storage medium, characterized in that The computer storage medium stores a computer software product, and the computer software product includes several instructions for enabling a computer device to execute the robotic arm path planning method based on adaptive adjustment of environmental parameters as described in any one of claims 1 to 7.

Images