An Industrial Robot Path Planning Method and Simulation Teaching Platform

Abstract

This invention relates to the field of industrial robot technology, and particularly to an industrial robot path planning method and simulation teaching platform. The method includes: determining a three-dimensional workspace and the starting and ending points of the path planning; establishing corresponding path trees with the starting and ending points as root nodes respectively; expanding the nodes of the two path trees using a target bias probability sampling method to obtain the expansion points for the current round; determining multiple associated points based on the expansion points obtained in the current round; performing collision detection on the expansion points and their corresponding associated points one by one according to a preset order; determining whether the two path trees can be merged; if so, merging the two path trees to obtain a collision-free path from the starting point to the ending point; otherwise, returning to the step of expanding the nodes of the two path trees using the target bias probability sampling method. This invention can improve the efficiency of industrial robot path planning and provide technical support for intelligent planning simulation interaction for industrial robots.

Description

Technical Field

[0001] This invention relates to the field of industrial robot technology, and in particular to an industrial robot path planning method, an industrial robot simulation teaching platform, electronic equipment, and storage medium. Background Technology

[0002] Industrial robots generally refer to multi-jointed manipulators or multi-degree-of-freedom machine devices used in industrial fields, such as welding robots. These robots move through space using an end effector (i.e., a welding torch) to automate the welding of workpieces. Industrial robots require programming and teaching before use. Currently, industrial robot programming and teaching are mainly divided into two methods: online programming and teaching, and offline programming and teaching. Compared to online programming and teaching, offline programming and teaching does not require large-scale shutdowns or production stoppages, has lower teaching costs, and because it primarily relies on computer graphics and other electronic technologies, users can obtain more comprehensive data during the teaching process, which helps to complete the teaching work more accurately.

[0003] While existing offline programming teaching platforms have initially achieved offline programming teaching, they are usually slow in path planning, still require manual assistance, have a low level of intelligence, and are limited in applicable scenarios. Summary of the Invention

[0004] To address at least some of the shortcomings mentioned above, embodiments of the present invention provide an industrial robot path planning method, an industrial robot simulation teaching platform, electronic devices, and storage media, which can improve the efficiency of industrial robot path planning and provide technical support for intelligent planning simulation interaction of industrial robots.

[0005] In a first aspect, embodiments of the present invention provide an industrial robot path planning method, comprising:

[0006] The three-dimensional workspace and the starting and ending points of the path planning are determined; the three-dimensional workspace is used to simulate the actual working environment of the industrial robot.

[0007] Establish corresponding path trees with the starting point and the ending point as root nodes respectively;

[0008] The expansion points for this round are obtained by expanding the nodes of the two path trees using the target bias probability sampling method.

[0009] Based on the expansion points obtained in this round, multiple corresponding associated points are determined;

[0010] According to the preset order, collision detection is performed on each of the expansion points and their corresponding associated points obtained in this round. If the expansion point or any associated point passes the collision detection, the expansion point or associated point is added as a new node to the corresponding path tree and the next step is executed. If the expansion point and each associated point fail the collision detection, the expansion point and each associated point obtained in this round are discarded and the next step is executed.

[0011] Determine whether the two path trees can be merged. If so, merge the two path trees to obtain a collision-free path from the starting point to the ending point. Otherwise, return to the step of expanding the nodes of the two path trees respectively by using the target bias probability sampling method.

[0012] Optionally, the step of expanding the nodes of the two path trees using the target bias probability sampling method includes performing the following steps on each of the two path trees:

[0013] Determine the probability of random sampling in this round;

[0014] If the random sampling probability is greater than the preset probability threshold, then starting from the current point, the latest node of another path tree is used as the target point and the step size is twice the preset step size to move towards the target point to generate an extension point.

[0015] If the random sampling probability is not greater than the preset probability threshold, then a random point is generated in the non-physical region of the three-dimensional workspace. Starting from the current point, the random point is used as the target point and the point is shifted towards the target point with a preset step size to generate an extension point.

[0016] Optionally, the determination of multiple associated points based on the expansion points obtained in this round includes:

[0017] If the random sampling probability is greater than the preset probability threshold when generating an extension point, then at least 3 companion points are determined at a distance of 1 preset step length from the current point, of which 1 companion point is located at the midpoint between the extension point and the current point, and the angle between the line connecting the remaining companion points and the current point and the line connecting the current point and the target point is a preset first angle value.

[0018] If the random sampling probability is not greater than a preset probability threshold when generating extension points, then at least two companion points are determined at a distance of 1 preset step length from the current point. The angle between the line connecting the companion point and the current point and the line connecting the current point and the target point is a preset first angle value.

[0019] Optionally, the first angle value ranges from 45° to 70°.

[0020] Optionally, the industrial robot path planning method further includes:

[0021] Optimize the collision-free path from the starting point to the ending point.

[0022] Secondly, embodiments of the present invention also provide an industrial robot simulation teaching platform, comprising:

[0023] The environment module is used to build or load the 3D workspace of industrial robots and display it in a visual form;

[0024] The teaching module is used to adjust the robot's pose within the three-dimensional workspace based on input commands;

[0025] The path planning module is used to obtain a collision-free path from the starting point to the ending point of the path planning using the industrial robot path planning method described in any embodiment of this specification.

[0026] The simulation demonstration module is used to perform a visual simulation demonstration of a defined path;

[0027] The output module is used to output the determined path to the industrial robot in a preset format;

[0028] The co-simulation module is used to pass the determined path to the RobotStudio software to obtain the simulation results of the path from the RobotStudio software.

[0029] Optionally, the path planning module is further configured to obtain a collision-free path from the starting point to the ending point of the path planning using one or more of the following methods: RRT method, RRT* method, GB-RRT method, bidirectional RRT method, and ant colony method, and to filter multiple collision-free paths from the starting point to the ending point of the path planning.

[0030] Optionally, the path planning module is further configured to optimize the determined collision-free path, including:

[0031] Smooth the determined collision-free path, and / or

[0032] Redundancy removal is performed on the determined collision-free path.

[0033] Thirdly, embodiments of the present invention also provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the industrial robot path planning method described in any embodiment of this specification.

[0034] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute the industrial robot path planning method described in any embodiment of this specification.

[0035] This invention provides an industrial robot path planning method, electronic device, and storage medium. When performing path planning, this invention expands bidirectionally from the starting point and the ending point using a target bias probability sampling method to improve path search efficiency. At the same time, it determines the corresponding associated points based on the expansion points and performs collision detection on the expansion points and each associated point to increase the probability of obtaining new nodes in each round of search, thereby further improving the path planning processing efficiency.

[0036] This invention also provides an industrial robot simulation teaching platform. This platform can display the three-dimensional workspace of the industrial robot in a visual form and adjust the posture of the industrial robot according to user input commands. The interaction method is simpler, and it uses the above-mentioned industrial robot path planning method to intelligently and automatically determine the path, resulting in higher processing efficiency. At the same time, it can perform visual simulation demonstration of the determined path so that users can intuitively determine the teaching results. It is applicable to different industrial robot working scenarios and has good versatility. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a flowchart of an industrial robot path planning method provided in an embodiment of the present invention;

[0039] Figures 2(a) and 2(b) are schematic diagrams showing the positional relationship between the extension points and the corresponding associated points in one embodiment of the present invention;

[0040] Figure 3(a) shows the path planning results of the RRT method in scenario 1;

[0041] Figure 3(b) shows the path planning results of the RRT* method in scenario 1;

[0042] Figure 3(c) shows the path planning results of the GB-RRT method in scenario 1;

[0043] Figure 3(d) shows the path planning results of the bidirectional RRT method in scenario 1;

[0044] Figure 3(e) shows the path planning results of the method of the present invention in scenario 1;

[0045] Figure 4(a) shows the path planning results of the RRT method in scenario 2;

[0046] Figure 4(b) shows the path planning results of the RRT* method in scenario 2;

[0047] Figure 4(c) shows the path planning results of the GB-RRT method in scenario 2;

[0048] Figure 4(d) shows the path planning results of the bidirectional RRT method in scenario 2;

[0049] Figure 4(e) shows the path planning results of the method of the present invention in scenario 2;

[0050] Figure 5(a) shows the average path planning time for each method to run 50 times in scenarios 1 and 2;

[0051] Figure 5(b) shows the average path length after each method is run 50 times in scenarios 1 and 2;

[0052] Figure 5(c) shows the average number of iterations for each method to run 50 times in scenarios 1 and 2;

[0053] Figure 6 This is a hardware architecture diagram of an electronic device provided in an embodiment of the present invention;

[0054] Figure 7 This is a structural diagram of an industrial robot simulation teaching platform provided in an embodiment of the present invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0056] As mentioned earlier, industrial robot programming teaching mainly falls into two categories: online programming teaching and offline programming teaching. Offline programming teaching involves constructing a 3D virtual environment of the entire robot's working scenario within the programming software, generating the robot's motion trajectory based on processing technology and other relevant requirements, simulating and adjusting it within the software, and finally generating the robot's execution program for transmission to the robot. Compared to online programming teaching, offline programming teaching is less expensive and allows for more precise teaching.

[0057] While existing offline programming teaching platforms have achieved basic offline programming teaching, their path planning speed is typically slow, and manual assistance is still required. Engineers often need to manually correct and adjust the platform based on the specific characteristics of different robots and workpieces to complete the teaching task. This results in low levels of intelligence and limited applicability. Therefore, it is necessary to propose a more efficient industrial robot path planning method and a more intelligent and universally applicable offline programming teaching technology for industrial robots.

[0058] Please refer to Figure 1 This invention provides an industrial robot path planning method, which includes:

[0059] Step 100: Determine the three-dimensional workspace of the industrial robot and the start and end points of the path planning;

[0060] The three-dimensional workspace is used to simulate the actual working environment of an industrial robot; the three-dimensional workspace typically includes one or more of the following: robot body model, end effector model, workpiece model, and obstacle model; obstacles are other physical objects besides the robot body, end effector, and workpiece.

[0061] Step 102: Establish corresponding path trees with the starting point and the ending point as root nodes respectively;

[0062] This step 102 establishes a two-path tree to enable bidirectional search and improve processing efficiency;

[0063] Step 104: Expand the nodes of the two path trees respectively using the target bias probability sampling method to obtain the expansion points for this round;

[0064] This step 104 is used to obtain the nodes determined in this round of expansion (or this round of search);

[0065] Step 106: Based on the expansion points obtained in this round, determine the corresponding multiple associated points;

[0066] Step 108: Perform collision detection on the expansion points and corresponding associated points obtained in this round one by one in a preset order;

[0067] If the extended point or any associated point passes the collision detection, then the extended point or associated point is added as a new node to the corresponding path tree, and the next step is executed.

[0068] If the expansion point and all associated points fail the collision detection, discard the expansion point and all associated points obtained in this round, and proceed to the next step;

[0069] Step 110: Determine whether the two path trees can be merged. If yes, merge the two path trees to obtain a collision-free path from the starting point to the ending point. Otherwise, return to step 104 to obtain a new round of expansion points. Through multiple iterations, a feasible path is obtained to achieve path planning.

[0070] This invention employs a target-biased probability sampling method to simultaneously expand the path planning from both the starting and ending points, thereby improving path search efficiency. Furthermore, it identifies multiple associated points for each expanded point obtained in each search round and performs collision detection on both the expanded points and each associated point. As a result, the probability of obtaining collision-free nodes in a single search round is significantly increased, which helps to reduce the number of search rounds (i.e., the number of iterations), thereby improving the computational efficiency of path planning and saving processing time.

[0071] Optionally, for step 104, "expanding nodes of the two path trees respectively using the target bias probability sampling method", the following steps are performed on the two path trees established in step 102 respectively:

[0072] Determine the probability of random sampling in this round;

[0073] If the determined random sampling probability is greater than the preset probability threshold, then starting from the current point (i.e. the latest node of the path tree currently being expanded), the latest node of another path tree is used as the target point and the expansion point is shifted towards the target point with a step size of 2 times the preset step size, thus generating an expansion point.

[0074] If the determined random sampling probability is not greater than a preset probability threshold, then a random point is generated in the non-physical region of the three-dimensional workspace. Starting from the current point, the generated random point is used as the target point and the point is shifted towards the target point with a preset step size to generate an extension point.

[0075] In the above embodiments, the target point can be determined by the following formula:

[0076]

[0077] Where, q point This represents the latest node in another path tree, updated as nodes in the other path tree are updated. q rand This refers to the random points generated in the non-physical regions (i.e., the spatial regions outside the areas occupied by the robot body model, end effector model, workpiece model, and obstacle model) within the three-dimensional workspace. P represents the random sampling probability, and P0 represents the probability threshold. The value of P0 is related to the complexity of the three-dimensional workspace. The lower the complexity of the three-dimensional workspace, the fewer the obstacles, and the larger the non-physical regions, the smaller the value of P0. Conversely, the higher the complexity of the three-dimensional workspace, the more obstacles, and the smaller the non-physical regions, the larger the value of P0, in order to achieve fast search.

[0078] The above embodiments expand the nodes of the two path trees respectively by using the target bias probability sampling method. When expanding towards the latest node of the other path tree, the expansion is carried out with a step size of 2 times the preset size, which can accelerate the two path trees to move closer to each other, so as to quickly determine the collision-free path from the starting point to the ending point and improve the efficiency of path planning.

[0079] It should be noted that if a collision-free path from the starting point to the ending point cannot be obtained after multiple iterations, the position of the latest node in the path tree should be adjusted, that is, some nodes should be deleted, the path should be reverted to the previously expanded nodes, and the path search should be performed again.

[0080] Furthermore, regarding step 106, "based on the expansion points obtained in this round, determine the corresponding multiple associated points," this includes:

[0081] If the random sampling probability P is greater than the preset probability threshold P0 when generating the extension point, then at least 3 companion points are determined at a distance of 1 preset step length from the current point, of which 1 companion point is located at the midpoint between the extension point and the current point, and the angle between the line connecting the remaining companion points and the current point and the line connecting the current point and the target point is a preset first angle value.

[0082] If the random sampling probability P is not greater than the preset probability threshold P0 when generating extension points, then at least two companion points are determined at a distance of 1 preset step length from the current point. The angle between the line connecting the companion point and the current point and the line connecting the current point and the target point is a preset first angle value.

[0083] Figures 2(a) and 2(b) are schematic diagrams showing the positional relationship between the extension point and the corresponding associated point in one embodiment of the present invention. If the random sampling probability P is greater than the probability threshold P0 when generating the extension point in step 104, as shown in Figure 2(a), let the starting point of the path planning be q. init The endpoint is q goal The current point of this path tree is q. near Then the extended point q1 is derived from the current point q. near Towards the latest node q of another path tree point The offset is obtained by 2 times the preset step size, at the extended point q1 and the current point q. near There is one associated point q2 at the midpoint, and two other associated points q3 and q4 are connected to the current point q. near The connection q3-q near ,q4-q near All lengths are 1 times the preset step size, and the included angle of the connecting lines is ∠q3q. near q1、∠q4q near The angle of q1 is always the first angle value. If the random sampling probability P is not greater than the probability threshold P0 when generating the extension point in step 104, as shown in Figure 2(b), the extension point q1 is generated from the current point q. nearTowards the generated random point q rand Offset by 1 times the preset step size, the two associated points q2 and q3 are obtained, along with the current point q. near The line connecting q2-q near ,q3-q near All lengths are 1 times the preset step size, and the angle between the connecting lines is ∠q2q. near q1、∠q3q near The angle of q1 is always the first angle value.

[0084] Using the above embodiments, multiple associated points can be quickly determined at the same time as the expansion point. The associated points are close to the expansion point and are associated with its location. Even if the expansion point fails the collision detection, the associated points may still pass the collision detection. In each round of search, in addition to the expansion point, the associated points also have the opportunity to be added to the path tree as new nodes. In this way, the probability of finding new nodes that can pass the collision detection is significantly improved, which helps to shorten the search rounds and thus improve the efficiency of path planning.

[0085] Furthermore, the first angle value is preferably in the range of 45° to 70°, and more preferably 60°.

[0086] If the first angle value is too small, the accompanying point will be too close to the expansion point. If the expansion point fails collision detection, the accompanying point will likely also fail, thus failing to effectively increase the probability of finding new nodes that pass collision detection in each round of search. If the first angle value is too large, the accompanying point may lose its tendency to move towards the target point, potentially causing path redundancy, i.e., taking detours and increasing processing time. Within the range of 45° to 70°, especially when the first angle value is 60°, the accompanying point is relatively close to the expansion point but still has a certain distance. When the expansion point fails collision detection, the accompanying point still has a high probability of passing. Furthermore, when the expansion point moves towards the target point, the corresponding accompanying point also tends to move closer to the target point. This increases the probability of finding new nodes and accelerates the merging of the two path trees. Too few accompanying points also hinder the improvement of the probability of finding new nodes, while too many may increase the computational burden. For cases where the random sampling probability P is not greater than the probability threshold P0, at least two associated points are determined. In the two-dimensional case, this means determining two associated points at 60° angles to the left and right of the extended point. For cases where the random sampling probability P is greater than the probability threshold P0, at least three associated points are determined. In the two-dimensional case, this means determining the midpoint and two associated points at 60° angles to the left and right of the midpoint.

[0087] Please see Figures 3(a) to 5(c)To verify the performance of the industrial robot path planning method provided by the present invention (hereinafter referred to as the present invention method or the improved RRT* method), the present invention sets up a scenario 1 representing a simple three-dimensional environment and a scenario 2 representing a complex three-dimensional environment, and runs the present invention method and existing RRT methods, RRT* methods, GB-RRT methods and bidirectional RRT methods in scenario 1 and scenario 2 respectively for path planning. Figures 3(a) to 3(e) The path planning results of the RRT method, RRT* method, GB-RRT method, bidirectional RRT method, and the method of this invention in scenario 1 are shown respectively. Figures 4(a) to 4(e) The path planning results of the RRT method, RRT* method, GB-RRT method, bidirectional RRT method and the method of the present invention in scenario 2 are shown respectively. In the example, the black sphere represents the set entity region (i.e. the region where the obstacle is located), the gray thin line represents the expanded path tree, i.e. the saved collision-free path, and the dark gray thick line represents a path from the start point to the end point found from all the input collision-free paths. Figures 5(a) to 5(c) The average path planning time, path length, and number of iterations for the RRT method, RRT* method, GB-RRT method, bidirectional RRT method, and the method of this invention, run 50 times in scenarios 1 and 2, are shown respectively. Figures 3(a) to 5(c) It can be seen that in the simpler scenario 1 and the more complex scenario 2, the method of the present invention (i.e. the improved RRT* method) has the minimum average computation time, path length and number of iterations, and the path planning performance is optimized.

[0088] Optionally, in step 108, "perform collision detection on each of the extended points and corresponding associated points obtained in this round," including:

[0089] The expansion points and their corresponding associated points obtained in this round are used as pre-sampling points and arranged in a preset order. In the preset order, the expansion point is before all associated points. If an associated point is at the midpoint between the expansion point and the current point, then the associated point is before all other associated points. The order of the remaining associated points is random.

[0090] In the order of arrangement, the inverse solution is calculated for each pre-sampling point. Based on the information of the entity region in the three-dimensional workspace (i.e. the area occupied by the robot body model, the end effector model, the workpiece model and the obstacle model), it is detected whether a collision occurs. If a collision occurs, it is considered that the collision detection fails; otherwise, it is considered that the collision detection passes.

[0091] Optionally, the industrial robot path planning method further includes:

[0092] Step 112: Optimize the collision-free path from the starting point to the ending point obtained in step 110.

[0093] While the industrial robot path obtained through the search may meet the work requirements, it may not be smooth enough, which is not conducive to the actual movement of the industrial robot body and end effector, or there may be redundancy in the path (i.e. taking too many detours). The above embodiments optimize the collision-free path obtained through the search, smooth the path, and / or remove redundant points from the path, which is beneficial to improving the robot's working efficiency.

[0094] like Figure 6 , Figure 7 As shown, this embodiment of the invention provides an industrial robot teaching platform, which can be implemented through software, hardware, or a combination of both. From a hardware perspective, as... Figure 6 The diagram shown is a hardware architecture diagram of an electronic device housing an industrial robot teaching platform provided in an embodiment of the present invention. (Except for...) Figure 6 In addition to the processor, memory, network interface, and non-volatile memory shown, the electronic device in the embodiment may also include other hardware, such as a forwarding chip responsible for processing packets. Taking software implementation as an example, such as... Figure 7 As shown, this teaching platform, as a logical device, is formed by the CPU of its host electronic device reading the corresponding computer program from the non-volatile memory into memory for execution. This embodiment provides an industrial robot teaching platform, including:

[0095] Environment module 701 is used to build or load the three-dimensional workspace of an industrial robot and display it in a visual form;

[0096] The teaching module 702 is used to adjust the robot's pose in the three-dimensional workspace based on input commands;

[0097] The path planning module 703 is used to obtain a collision-free path from the starting point to the ending point of the path planning by means of the industrial robot path planning method described in any of the above embodiments.

[0098] The simulation demonstration module 704 is used to perform a visual simulation demonstration of a defined path;

[0099] Output module 705 is used to output the determined path to the industrial robot in a preset format;

[0100] The co-simulation module 706 is used to pass the determined path to the RobotStudio software to obtain the simulation results of the path by the RobotStudio software.

[0101] The industrial robot teaching platform provided in the above embodiments automatically plans paths using an intelligent path planning method, effectively improving teaching efficiency and the accuracy of teaching results. Furthermore, the environment module 701 displays the three-dimensional workspace of the industrial robot in a visual form, and the teaching module 702 is used for parameter adjustment, which can adjust the robot's pose within the three-dimensional workspace. The robot's control method is intuitive and can meet the teaching needs of common six-axis robots in spot welding, arc welding, and other operation scenarios. At the same time, it simplifies the teaching method, has strong versatility, and helps to shorten the robot teaching cycle.

[0102] Optionally, the environment module 701 constructs or loads the three-dimensional workspace of the industrial robot, including performing the following operations:

[0103] Determine the models of each entity in the industrial robot's working scenario, including the robot body model, the end effector model, and the workpiece model;

[0104] Based on the poses of the robot body and the end effector in the industrial robot working scenario, the poses of the robot body model and the end effector model in the three-dimensional workspace are determined by the DH parameter method.

[0105] Based on the relative positional relationship between the workpiece and the industrial robot in the industrial robot's working scenario and the pose of the robot body model in the three-dimensional workspace, the pose of the workpiece model in the three-dimensional workspace is determined.

[0106] Optionally, if other entities, such as obstacles, exist in the industrial robot's working environment, the environment module 701 constructing or loading the industrial robot's three-dimensional workspace also includes performing the following operations:

[0107] Based on the relative positional relationship between obstacles and industrial robots in the industrial robot's working scenario and the pose of the robot body model in the three-dimensional workspace, the pose of the obstacle model in the three-dimensional workspace is determined.

[0108] Commonly used industrial robot models such as ABB and KUKA can be directly imported into the 3D workspace for use. First, the robot's pose is determined through calibration technology, and then other entities are determined based on their relative positions. This helps ensure that the positional relationships of each entity's model match in the offline programming simulation environment (i.e., the 3D workspace) and the actual work scene. The robot model imported into the platform preferably uses a relatively simple STL file format. The environment module 701 is used to provide a more realistic and three-dimensional display, while the teaching module 702 displays various robot parameters on the teach pendant interface, allowing users to adjust the robot's joint angles, specific positions, and orientation. This reduces the difficulty of importing self-designed workpieces and robot models into the platform, improves the platform's applicability to new workpieces and robots, enables fine-tuning of the robot's pose, and makes the robot teaching method and effect more flexible, lowering the technical threshold for offline teaching technology.

[0109] Optionally, the path planning module 703 is further configured to obtain a collision-free path from the starting point to the ending point of the path planning using one or more of the RRT method, RRT* method, GB-RRT method, bidirectional RRT method, and ant colony method, and to filter the multiple collision-free paths obtained from the starting point to the ending point of the path planning.

[0110] In the above embodiments, the platform integrates multiple path planning methods, making it convenient for users to compare the advantages and disadvantages of different path planning methods and thus select the most suitable path planning result.

[0111] Furthermore, the path planning module 703 filters multiple collision-free paths from the starting point to the ending point of the path planning, including performing the following operations:

[0112] Based on the input instructions, filter multiple collision-free paths from the start point to the end point of the path planning; and / or

[0113] Based on preset screening criteria, multiple collision-free paths from the starting point to the destination in path planning are selected. Screening criteria can include minimum computation time, minimum path length, etc.

[0114] By filtering multiple collision-free paths from the starting point to the destination of the path planning based on input commands, the solution that the user is interested in can be obtained. By filtering multiple collision-free paths from the starting point to the destination of the path planning based on preset filtering criteria, the optimal solution can be automatically selected.

[0115] Furthermore, the path planning module 703 is also used to optimize the determined collision-free path, including:

[0116] Smooth the determined collision-free path, and / or

[0117] Redundancy removal is performed on the determined collision-free path.

[0118] Smoothing processes can improve path quality, ensuring smooth robot motion along the overall path and thus increasing actual operational efficiency. Redundant point removal can shorten path length, reducing detours and further improving efficiency. Optimizing a defined collision-free path can utilize existing techniques, such as redundant point removal using a greedy algorithm to extract and detect key points.

[0119] Optionally, the industrial robot teaching platform further includes a path correction module for correcting the current path based on input commands.

[0120] Using the above embodiments, after the determined path is visualized and simulated by the simulation demonstration module 704, if the user is not satisfied with the current solution, they can continue to input commands to modify the path and make adjustments based on the current path, which is more efficient and more intuitive.

[0121] Optionally, the output module 705 outputs the determined path to the industrial robot in a preset format, including extracting the point information of the path, converting it into a robot programming language format, and outputting it to the industrial robot.

[0122] RobotStudio software, a commonly used auxiliary software in the field of industrial robotics, can simulate the execution path of industrial robots. The co-simulation module 706 is used to connect with the existing RobotStudio software, converting the extracted path point information into program files used by RobotStudio software, facilitating the simulation of the determined path (i.e., the teaching result) in RobotStudio software.

[0123] Furthermore, the platform's path correction module can also correct the current path based on the co-simulation results (i.e., the path simulation results from RobotStudio software), achieving dynamic adjustment. The design of the co-simulation module 706 verifies the feasibility of inter-platform co-simulation and facilitates dynamic adjustment of offline teaching results.

[0124] It is understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on an industrial robot teaching platform. In other embodiments of the present invention, an industrial robot teaching platform may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0125] The information interaction and execution process between the modules in the above-mentioned device are based on the same concept as the method embodiment of the present invention, and the specific details can be found in the description of the method embodiment of the present invention, and will not be repeated here.

[0126] This invention also provides an electronic device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements an industrial robot path planning method according to any embodiment of this invention.

[0127] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform an industrial robot path planning method according to any embodiment of this invention.

[0128] Specifically, a system or apparatus equipped with a storage medium may be provided, on which software program code implementing the functions of any of the embodiments described above is stored, and the computer (or CPU or MPU) of the system or apparatus may read and execute the program code stored in the storage medium.

[0129] In this case, the program code read from the storage medium can itself implement the function of any of the above embodiments, and therefore the program code and the storage medium storing the program code constitute part of the present invention.

[0130] Examples of storage media used to provide program code include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, program code can be downloaded from a server computer via a communication network.

[0131] Furthermore, it should be clear that not only can the program code read by the computer be executed, but also the operating system or other components operating on the computer can be instructed based on the program code to perform some or all of the actual operations, thereby realizing the function of any of the embodiments described above.

[0132] Furthermore, it is understood that the program code read from the storage medium is written to the memory set in the expansion board inserted into the computer or to the memory set in the expansion module connected to the computer. Then, based on the instructions of the program code, the CPU or other components installed on the expansion board or expansion module execute some and all of the actual operations, thereby realizing the function of any of the above embodiments.

[0133] The embodiments of the present invention have at least the following beneficial effects:

[0134] 1. In one embodiment of the present invention, an industrial robot path planning method is provided. The method expands the path from the starting point and the ending point in both directions by using the target bias probability sampling method to improve the path search efficiency. At the same time, in each round of search, the corresponding associated points are determined according to the expansion points, and collision detection is performed on the expansion points and each associated point to increase the probability of obtaining new nodes of the collision-free path in each round of search, thereby further improving the path planning processing efficiency.

[0135] 2. In one embodiment of the present invention, an industrial robot teaching platform is provided. The platform adopts an intelligent path planning method to improve teaching efficiency and the accuracy of teaching results. It displays the three-dimensional workspace of the industrial robot in a visual form and makes corrections according to user input instructions. The interaction method is simpler and the robot control method is more intuitive. It can also realize visual simulation demonstration and co-simulation, which facilitates dynamic adjustment of offline teaching results.

[0136] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0137] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as ROM, RAM, magnetic disk, or optical disk.

[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A path planning method for an industrial robot, characterized in that, Includes the following steps: The three-dimensional workspace and the starting and ending points of the path planning are determined; the three-dimensional workspace is used to simulate the actual working environment of the industrial robot. Establish corresponding path trees with the starting point and the ending point as root nodes respectively; The expansion points for this round are obtained by expanding the nodes of the two path trees using the target bias probability sampling method. Based on the expansion points obtained in this round, multiple corresponding associated points are determined; According to the preset order, collision detection is performed on each of the expansion points and their corresponding associated points obtained in this round. If the expansion point or any associated point passes the collision detection, the expansion point or associated point is added as a new node to the corresponding path tree and the next step is executed. If the expansion point and each associated point fail the collision detection, the expansion point and each associated point obtained in this round are discarded and the next step is executed. Determine whether the two path trees can be merged. If so, merge the two path trees to obtain a collision-free path from the starting point to the ending point. Otherwise, return to the step of expanding the nodes of the two path trees respectively by using the target bias probability sampling method. The node expansion of the two path trees using the target bias probability sampling method includes performing the following steps on each path tree: Determine the probability of random sampling in this round; If the random sampling probability is greater than the preset probability threshold, then starting from the current point, the latest node of another path tree is used as the target point and the step size is twice the preset step size to move towards the target point to generate an extension point. If the random sampling probability is not greater than the preset probability threshold, then a random point is generated in the non-physical region of the three-dimensional workspace. Starting from the current point, the random point is used as the target point and the point is shifted towards the target point with a preset step size to generate an extension point.

2. The industrial robot path planning method according to claim 1, characterized in that, Based on the expansion points obtained in this round, multiple associated points are determined, including: If the random sampling probability is greater than the preset probability threshold when generating an extension point, then at least 3 companion points are determined at a distance of 1 preset step length from the current point, of which 1 companion point is located at the midpoint between the extension point and the current point, and the angle between the line connecting the remaining companion points and the current point and the line connecting the current point and the target point is a preset first angle value. If the random sampling probability is not greater than a preset probability threshold when generating extension points, then at least two companion points are determined at a distance of 1 preset step length from the current point. The angle between the line connecting the companion point and the current point and the line connecting the current point and the target point is a preset first angle value.

3. The industrial robot path planning method according to claim 2, characterized in that, The first angle value ranges from 45° to 70°.

4. The industrial robot path planning method according to claim 1, characterized in that, Also includes: Optimize the collision-free path from the starting point to the ending point.

5. An industrial robot teaching platform, characterized in that, include: The environment module is used to build or load the 3D workspace of industrial robots and display it in a visual form; The teaching module is used to adjust the robot's pose within the three-dimensional workspace based on input commands; A path planning module is used to obtain a collision-free path from the starting point to the ending point of the path planning using the industrial robot path planning method as described in any one of claims 1-4. The simulation demonstration module is used to perform a visual simulation demonstration of a defined path; The output module is used to output the determined path to the industrial robot in a preset format; The co-simulation module is used to pass the determined path to the RobotStudio software to obtain the simulation results of the path from the RobotStudio software.

6. The industrial robot teaching platform according to claim 5, characterized in that, The path planning module is also used to obtain a collision-free path from the starting point to the ending point of the path planning using one or more of the following methods: RRT method, RRT* method, GB-RRT method, bidirectional RRT method, and ant colony method, and to filter multiple collision-free paths from the starting point to the ending point of the path planning.

7. The industrial robot teaching platform according to any one of claims 5 or 6, characterized in that, The path planning module is also used to optimize the determined collision-free path, including: Smooth the determined collision-free path, and / or Redundancy removal is performed on the determined collision-free path.

8. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the industrial robot path planning method as described in any one of claims 1-4.

9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed in the computer, it causes the computer to perform the industrial robot path planning method according to any one of claims 1-4.

Images