Six-axis Robot Obstacle Avoidance Path Planning Method and System Based on Adam Optimization Algorithm

Abstract

A method and system for obstacle avoidance path planning of a six-axis robot based on the adam optimization algorithm, wherein the method includes: obtaining path endpoints corresponding to the current six-axis robot, wherein the path endpoints include a starting joint value and an ending joint value; performing uniform interpolation between the path endpoints to obtain path key points; sampling based on a preset step size, and combining the path endpoints and the path key points to obtain the number of collision voxels; obtaining a loss function value based on the number of collision voxels, and using the adam optimization algorithm to iteratively solve the loss function to obtain a planned path when the loss is zero, wherein the number of collisions on the planned path is the least. The present invention uses the adam algorithm to perform gradient descent optimization on the running path, and even in the face of a complex application environment, it can achieve an obstacle avoidance effect in the path planning of the robot, thereby improving the running efficiency of the robot and reducing the collision rate on the traveling path.

Description

Technical Field

[0001] The present invention relates to the field of robot operation technology, and more specifically, to a six-axis robot obstacle avoidance path planning method and system based on an Adam optimization algorithm. Background Art

[0002] In robotics, many methods have been applied to solve robot kinematics and collaboration problems. Among them, traditional methods include heuristic search-based algorithms and optimization algorithms. In recent years, optimization algorithms based on gradient descent have gradually attracted attention, especially when dealing with complex nonlinear problems.

[0003] During the process, the robot's obstacle avoidance path is particularly important. The current mainstream obstacle avoidance planning scheme is the RRT+APF artificial potential field method and the heuristic space search method, such as the particle swarm algorithm. However, the artificial field potential often faces the problem of being unable to solve for a long time or path oscillation in a complex environment, and the heuristic search method has a too large search space, resulting in a long time of failure to converge. This problem has yet to be solved. Summary of the invention

[0004] The purpose of the present invention is to provide a six-axis robot obstacle avoidance path planning method and system based on the Adam optimization algorithm, which uses the Adam algorithm to perform gradient descent optimization on the running path. Even in the face of complex application environments, the obstacle avoidance effect can be achieved in the robot's path planning, thereby improving the robot's operating efficiency and reducing the collision rate on the travel path.

[0005] A first aspect of the present invention provides a six-axis robot obstacle avoidance path planning method based on the Adam optimization algorithm, comprising the following steps:

[0006] Obtain the path endpoint corresponding to the current six-axis robot, wherein the path endpoint includes a start point joint value and an end point joint value;

[0007] Uniformly interpolating between the path endpoints to obtain path key points;

[0008] Sampling is performed based on a preset step size, and the number of collision voxels is obtained by combining the path endpoints and the path key points;

[0009] The loss function value is obtained based on the number of collision voxels, and the loss function is iteratively solved using the Adam optimization algorithm to obtain the planned path when the loss is zero.

[0010] In this solution, obtaining the path endpoint corresponding to the current six-axis robot specifically includes:

[0011] The path endpoint is obtained based on the current working coordinate system of the six-axis robot, where:

[0012] Obtain the starting joint values based on the current position coordinates of the robot ;

[0013] Obtain the ending joint values based on the position coordinates of the forward destination of the robot .

[0014] In this solution, performing uniform interpolation between the path endpoints to obtain path key points specifically includes:

[0015] Performing N uniform interpolations between the starting joint values and the ending joint values to obtain the path key points, where the path key points are the joint values for the robot to execute the target instruction;

[0016] The calculation formula for obtaining N path key points is as follows:

[0017] ;

[0018] Where, is the joint value corresponding to the th path key point, is the interpolation function, is the number of path key points.

[0019] In this solution, the calculation expression of the interpolation function is as follows: , where, is the actual calculation parameter value.

[0020] In this solution, sampling based on a preset step size and combining the path endpoints and the path key points to obtain the number of collision voxels specifically includes:

[0021] Sampling based on a preset step size and obtaining a loss value based on the number of collision voxels, where the loss value has the following calculation formula:

[0022] ;

[0023] Where, represents the number of samples taken between every two key path points, represents the joint space distance for solving joint and joint , is the number of collision voxels in the collision area of the six-axis robot joint value J with the environment, is the path key point, is the number of path key points.

[0024] In this solution, the method further includes deleting redundant points on the planned path. Among them, if the loss is still zero when deleting the th path key point, then the current th path key point is taken as the redundant point and deleted.

[0025] The second aspect of the present invention further provides a six-axis robot obstacle avoidance path planning system based on the adam optimization algorithm, including a memory and a processor. The memory includes a six-axis robot obstacle avoidance path planning method program based on the adam optimization algorithm. When the six-axis robot obstacle avoidance path planning method program based on the adam optimization algorithm is executed by the processor, the following steps are implemented:

[0026] Obtain the path end points corresponding to the current six-axis robot, where the path end points include a starting joint value and an ending joint value;

[0027] Perform uniform interpolation between the path end points to obtain path key points;

[0028] Sample based on a preset step size, and combine the path end points and the path key points to obtain the number of collision voxels;

[0029] Obtain a loss function value based on the number of collision voxels, and use the adam optimization algorithm to iteratively solve the loss function to obtain a planned path when the loss is zero.

[0030] In this solution, the obtaining of the path end points corresponding to the current six-axis robot specifically includes:

[0031] Obtain the path end points based on the working coordinate system where the current six-axis robot is located. Among them,

[0032] Obtain the starting joint value based on the current position coordinates of the robot ;

[0033] Obtain the ending joint value based on the position coordinates of the robot's forward destination .

[0034] In this solution, the performing of uniform interpolation between the path end points to obtain path key points specifically includes:

[0035] Perform N uniform interpolations between the starting joint value and the ending joint value to obtain the path key points, where the path key points are the joint values for the robot to execute the target instruction;

[0036] The calculation formula for obtaining N path key points is as follows:

[0037] ;

[0038] Among them, is the joint value corresponding to the th path key point, is the interpolation function, is the number of path key points.

[0039] In this solution, the interpolation function has the following calculation expression: , where is the actual calculation parameter value.

[0040] In this solution, sampling based on a preset step size and combining the path endpoints and the path key points to obtain the number of collision voxels specifically includes:

[0041] Sampling based on a preset step size and obtaining a loss value based on the number of collision voxels, where the loss value has the following calculation formula:

[0042] ;

[0043] where represents the number of samplings between every two key path points, represents the joint space distance for solving joint and joint , is the number of collision voxels in the collision area when the joint value J of the six-axis robot and the environment, is the path key point, is the number of path key points.

[0044] In this solution, the method further includes deleting redundant points on the planned path, where if the loss is still zero when deleting the th path key point, then the current th path key point is deleted as the redundant point.

[0045] The third aspect of the present invention provides a computer-readable storage medium, which includes a program of a six-axis robot obstacle avoidance path planning method based on the adam optimization algorithm for a machine. When the program of the six-axis robot obstacle avoidance path planning method based on the adam optimization algorithm is executed by a processor, the steps of a six-axis robot obstacle avoidance path planning method as described in any one of the above are implemented.

[0046] A method and system for obstacle avoidance path planning of a six-axis robot based on the adam optimization algorithm disclosed in the present invention utilize the adam algorithm to optimize the gradient descent of the running path. Even in the face of a complex application environment, the obstacle avoidance effect can be achieved in the path planning of the robot, thereby improving the running efficiency of the robot and reducing the collision rate on the traveling path. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 The flowchart of a method for obstacle avoidance path planning of a six-axis robot based on the adam optimization algorithm according to the present invention is shown;

[0048] Figure 2 The schematic diagram of solving the planned path in a method for obstacle avoidance path planning of a six-axis robot based on the adam optimization algorithm according to the present invention is shown;

[0049] Figure 3 The block diagram of a system for obstacle avoidance path planning of a six-axis robot based on the adam optimization algorithm according to the present invention is shown. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] In order to more clearly understand the above objects, features and advantages of the present invention, the present invention will be further described in detail below with reference to the drawings and specific embodiments. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

[0051] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0052] Figure 1 The flowchart of a method for obstacle avoidance path planning of a six-axis robot based on the adam optimization algorithm according to the present application is shown.

[0053] As Figure 1 shown, the present application discloses a method for obstacle avoidance path planning of a six-axis robot based on the adam optimization algorithm, including the following steps:

[0054] S102, obtaining the path end points corresponding to the current six-axis robot, wherein the path end points include the starting joint values and the ending joint values;

[0055] S104, performing uniform interpolation between the path end points to obtain path key points;

[0056] S106, sampling based on a preset step size, and combining the path end points and the path key points to obtain the number of collision voxels;

[0057] S108. Obtain the loss function value based on the number of collision voxels, and use the adam optimization algorithm to iteratively solve the loss function to obtain the planned path when the loss is zero.

[0058] It should be noted that in this embodiment, the present application proposes an obstacle avoidance planning algorithm that directly performs gradient descent optimization on the motion path. Even in the face of a complex environment, it can give a sub-optimal solution. The Adam optimization algorithm used, as a gradient descent algorithm with an adaptive learning rate, has not been widely studied in the application of robot path planning. However, its ability to converge quickly and perform global search is very suitable for dealing with complex obstacle avoidance problems.

[0059] Furthermore, by solving the given starting joint values and ending joint values, a collision-free path is calculated. The core lies in using N evenly interpolated points between the starting joint values and ending joint values as path key points. The obtained path key points are the joint values for the robot to execute the target instructions. The target instructions include PTP instructions. Among them, PTP (Precision Time Protocol) is a network protocol designed to provide high-precision time synchronization. A path consists of multiple PTP instructions and is sampled at a certain step size. The total number of voxels collided by the robot during the process of reaching the end point from the starting point along this path is used as the loss function value to solve the loss function using the adam optimization algorithm. When the loss value is equal to 0, it is considered that the function has been solved. Even when reaching the upper limit of the maximum number of iterations, a path with as few collisions as possible can be obtained as the planned path. Correspondingly, the number of collisions on the planned path is the least.

[0060] Specifically, in the actual operation process, first obtain the path end points corresponding to the current six-axis robot. Among them, the path end points include the starting joint values and the ending joint values. Then, perform uniform interpolation between the path end points to obtain path key points. Then, sample based on a preset step size, and combine the path end points and the path key points to obtain the number of collision voxels. After obtaining the number of collision voxels, the loss function value can be obtained based on the number of collision voxels, and the adam optimization algorithm is used to iteratively solve the loss function to obtain the planned path when the loss is zero. The specific step details will be described in detail in the subsequent specification.

[0061] According to the embodiment of the present invention, the obtaining of the path end points corresponding to the current six-axis robot specifically includes:

[0062] Obtain the path end points based on the working coordinate system where the current six-axis robot is located. Among them,

[0063] Obtain the starting joint value based on the current position coordinates of the robot ;

[0064] Obtain the end joint value based on the position coordinates of the forward destination of the robot 。

[0065] It should be noted that in this embodiment, it specifically describes how to obtain the path endpoints. Among them, the path endpoints include a starting point and an ending point, which correspond to the starting joint value and the ending joint value on the robot. Specifically, obtain the path endpoints based on the working coordinate system where the current six-axis robot is located. The working coordinate system is the space coordinate system in the current robot working environment, so as to obtain the starting joint value based on the current position coordinates of the robot , and obtain the end joint value based on the position coordinates of the forward destination of the robot 。

[0066] According to the embodiment of the present invention, performing uniform interpolation between the path endpoints to obtain path key points specifically includes:

[0067] Perform N uniform interpolations between the starting joint value and the ending joint value to obtain the path key points, where the path key points are the joint values for the robot to execute the PTP instruction;

[0068] The calculation formula for obtaining N path key points is as follows:

[0069] ;

[0070] Among them, is the joint value corresponding to the th path key point, is the interpolation function, is the number of path key points.

[0071] It should be noted that in this embodiment, it specifically describes how to obtain the path key points. Among them, perform N uniform interpolations between the starting joint value and the ending joint value to obtain the path key points. Correspondingly, the obtained path key points are the joint values for the robot to execute the PTP instruction. A path is composed of multiple PTP instructions, so that the final planned path can be obtained by selecting different path key points. Specifically, the calculation formula for obtaining N path key points is as follows: ; Among them, is the joint value corresponding to the th path key point, is the interpolation function, is the number of path key points. Through the above calculations, the joint values corresponding to different path key points can be obtained, which is convenient for subsequent solving of the spatial distance between two joints.

[0072] According to the embodiment of the present invention, the interpolation function The calculation expression is as follows: , where is the actual calculation parameter value.

[0073] It should be noted that in this embodiment, the Lerp interpolation function is a commonly used function in Unity for linear interpolation between two numerical values. Its function is to interpolate between two values and return a numerical value between these two values. The full name of the Lerp function is Linear Interpolation, that is, linear interpolation. The specific expression is , where is the actual calculation parameter value. Since it is used as an application in this embodiment, the specific process of the interpolation function will not be elaborated here.

[0074] According to the embodiment of the present invention, sampling based on a preset step size and combining the path endpoints and the path key points to obtain the number of collision voxels specifically includes:

[0075] Sampling based on a preset step size and obtaining a loss value based on the number of collision voxels, where the loss value has the following calculation formula:

[0076] ;

[0077] where represents the number of samples taken between every two key path points, represents the solution of the joint and the joint in the joint space distance, is the number of collision voxels in the collision area when the six-axis robot joint value is J, is the path key point, is the number of path key points.

[0078] It should be noted that in this embodiment, the preset step size is a fixed step size, usually taking "0.2". When applying, first sample based on the preset step size and obtain the loss value of the current path by finding the number of voxels collided in each sample. Among them, the loss value has the following calculation formula: , where represents the number of samples taken between every two key path points, represents the solution of the joint and the joint in the joint space distance, is the number of collision voxels in the collision area when the six-axis robot joint value is J. In this embodiment, the voxel accuracy adopted is "5mm", is the path key point, is the number of path key points. In addition, the collision detection algorithm and the collision voxel number solving algorithm are also applied in this embodiment, so the specific application process will not be elaborated.

[0079] Further, during application, the input feature is the joint values of N key points, that is, an N * 6-dimensional vector, and the loss value of the loss function uses the adam optimization algorithm to numerically solve the function gradient for gradient descent, and converges when the loss equals "0". At this time, the final planned path is obtained. Among them, in the actual calculation process, there is a situation where the loss is not equal to zero. At this time, when the maximum iteration number limit is reached, the corresponding planned path is also solved. Correspondingly, the number of collisions on the obtained planned path is still the least.

[0080] According to the embodiment of the present invention, the method further includes deleting redundant points on the planned path. Among them, if when deleting the th path key point, the loss is still zero, then the current th path key point is deleted as the redundant point.

[0081] It should be noted that in this embodiment, referring to Figure 2 , it shows a schematic diagram of solving the planned path using the adam optimizer (optimization algorithm). Among them, the robot is first initialized, specifically including initializing the feature vector and the environmental voxels, and then the adam optimizer is used for updating until the loss equals zero, or the gradient equals zero, or the iteration number reaches the upper limit to end the optimization, and the redundant points on the current planned path are deleted to complete the solution to obtain the final planned path. Specifically, the specific judgment process for deleting the redundant points on the planned path is as follows: if when deleting the th path key point, the loss is still zero, then the current th path key point is deleted as the redundant point, which can reduce unnecessary travel without affecting path optimization.

[0082] Figure 3 shows a block diagram of a six-axis robot obstacle avoidance path planning system based on the adam optimization algorithm of the present invention.

[0083] As Figure 3 shown, the present invention discloses a six-axis robot obstacle avoidance path planning system based on the adam optimization algorithm, including a memory and a processor. The memory includes a six-axis robot obstacle avoidance path planning method program based on the adam optimization algorithm. When the six-axis robot obstacle avoidance path planning method program based on the adam optimization algorithm is executed by the processor, the following steps are implemented:

[0084] Obtain the path endpoints corresponding to the current six-axis robot, where the path endpoints include the starting joint values and the ending joint values;

[0085] Perform uniform interpolation between the path endpoints to obtain path key points;

[0086] Sample based on a preset step size, and combine the path endpoints and the path key points to obtain the number of collision voxels;

[0087] Obtain the loss function value based on the number of collision voxels, and use the adam optimization algorithm to iteratively solve the loss function to obtain the planned path when the loss is zero.

[0088] It should be noted that in this embodiment, the present application proposes an obstacle avoidance planning algorithm that directly performs gradient descent optimization on the motion path. Even in the face of a complex environment, it can give a sub-optimal solution. The Adam optimization algorithm used, as a gradient descent algorithm with an adaptive learning rate, has not been widely studied in the application of robot path planning. However, its ability to converge quickly and perform global search is very suitable for dealing with complex obstacle avoidance problems.

[0089] Furthermore, by solving the given starting joint values and ending joint values, a collision-free path is calculated. The core is to perform N uniform interpolation points between the starting joint values and the ending joint values as path key points. The obtained path key points are the joint values for the robot to execute PTP instructions. A path consists of multiple PTP instructions and is sampled at a certain step size. The total number of voxels collided by the robot during the process of reaching the end point from the starting point through this path is used as the loss function value. The adam optimization algorithm is used to solve the loss function. When the loss value is equal to 0, it is considered that the function has been solved. Even when the maximum number of iteration upper limits is reached, a path with as few collisions as possible can be obtained as the planned path. Correspondingly, the number of collisions on the planned path is the least.

[0090] Specifically, in the actual operation process, first obtain the path endpoints corresponding to the current six-axis robot, where the path endpoints include the starting joint values and the ending joint values, so as to perform uniform interpolation between the path endpoints to obtain path key points. Then sample based on a preset step size, and combine the path endpoints and the path key points to obtain the number of collision voxels. After obtaining the number of collision voxels, the loss function value can be obtained based on the number of collision voxels, and the adam optimization algorithm is used to iteratively solve the loss function to obtain the planned path when the loss is zero. The specific step details will be described in detail in the subsequent specification.

[0091] According to an embodiment of the present invention, the obtaining of the path endpoints corresponding to the current six-axis robot specifically includes:

[0092] The path endpoints are obtained based on the working coordinate system where the current six-axis robot is located. Among them,

[0093] the starting joint values are obtained based on the current position coordinates of the robot ;

[0094] the ending joint values are obtained based on the position coordinates of the forward destination of the robot .

[0095] It should be noted that in this embodiment, it specifically illustrates how to obtain the path endpoints. Among them, the path endpoints include a starting point and an ending point, which correspond to the starting joint values and ending joint values on the robot. Specifically, the path endpoints are obtained based on the working coordinate system where the current six-axis robot is located. The working coordinate system is the space coordinate system in the current robot working environment, so that the starting joint values are obtained based on the current position coordinates of the robot , and the ending joint values are obtained based on the position coordinates of the forward destination of the robot .

[0096] According to the embodiment of the present invention, performing uniform interpolation between the path endpoints to obtain path key points specifically includes:

[0097] Performing N uniform interpolations between the starting joint values and the ending joint values to obtain the path key points, where the path key points are the joint values for the robot to execute the PTP instruction;

[0098] The calculation formula for obtaining N path key points is as follows:

[0099] ;

[0100] Among them, is the joint value corresponding to the th path key point, is the interpolation function, is the number of path key points.

[0101] It should be noted that in this embodiment, it specifically illustrates how to obtain the path key points. Among them, N uniform interpolations are performed between the starting joint values and the ending joint values to obtain the path key points. Correspondingly, the obtained path key points are the joint values for the robot to execute the PTP instruction. A path is composed of multiple PTP instructions, so that the final planned path can be obtained by selecting at different path key points. Specifically, the calculation formula for obtaining N path key points is as follows: ; Among them, is the joint value corresponding to the th path key point, is the interpolation function, is the number of path key points. Through the above calculations, the joint values corresponding to different path key points can be obtained, which is convenient for subsequent solving of the spatial distance between two joints.

[0102] According to an embodiment of the present invention, the interpolation function has the following calculation expression: , where is the actual calculation parameter value.

[0103] It should be noted that in this embodiment, the Lerp interpolation function is a commonly used function in Unity for linear interpolation between two numerical values. Its function is to interpolate between two values and return a numerical value between these two values. The full name of the Lerp function is Linear Interpolation, that is, linear interpolation. The specific expression is , where is the actual calculation parameter value. Since it is used as an application in this embodiment, the specific process of the interpolation function will not be elaborated here.

[0104] According to an embodiment of the present invention, sampling based on a preset step size and combining the path endpoints and the path key points to obtain the number of collision voxels specifically includes:

[0105] Sampling based on a preset step size and obtaining a loss value based on the number of collision voxels, where the loss value has the following calculation formula:

[0106] ;

[0107] where represents the number of samples taken between every two key path points, represents solving the joint and the joint of the joint space distance, is the number of collision voxels in the collision area when the joint value J of the six-axis robot and the environment, is the path key point, is the number of path key points.

[0108] It should be noted that in this embodiment, the preset step size is a fixed step size, usually taking "0.2". When applying, first sample based on the preset step size, and find the number of voxels collided in each sample to obtain the loss value of the current path. Among them, the loss value has the following calculation formula: , where represents the number of samples taken between every two key path points, represents solving the joint and the joints in the joint space distance, is to calculate the number of collision voxels in the collision area when the six-axis robot joint value is J. In this embodiment, the voxel accuracy adopted is "5mm", is the key point of the path, is the number of key points of the path. In addition, the collision detection algorithm and the collision voxel number solving algorithm are also applied in this embodiment. Therefore, the specific application process will not be elaborated.

[0109] Furthermore, during application, the input feature is the joint values of N key points, that is, an N*6-dimensional vector, and the loss value of the loss function uses the adam optimization algorithm to numerically solve the function gradient for gradient descent. When the loss equals "0", it converges, and at this time, the final planned path is obtained. Among them, in the actual calculation process, there is a situation where the loss is not equal to zero. At this time, when the maximum number of iteration upper limits is reached, the corresponding planned path is also solved. Correspondingly, the number of collisions on the obtained planned path is still the least.

[0110] According to the embodiment of the present invention, the method further includes deleting redundant points on the planned path. Among them, if when deleting the th key point of the path, the loss is still zero, then the current th key point of the path is taken as the redundant point for deletion.

[0111] It should be noted that in this embodiment, referring to Figure 2 , it shows a schematic diagram of solving the planned path using the adam optimizer (optimization algorithm). Among them, the robot is first initialized, specifically including initializing the feature vector and the environmental voxels, and then the adam optimizer is used for update until the loss equals zero, or the gradient equals zero, or the number of iterations reaches the upper limit to end the optimization, and the redundant points on the current planned path are deleted to complete the solution to obtain the final planned path. Specifically, the specific judgment process for deleting the redundant points on the planned path is as follows: If when deleting the th key point of the path, the loss is still zero, then the current th key point of the path is taken as the redundant point for deletion, which can reduce unnecessary travel while not affecting path optimization.

[0112] In a third aspect of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium includes a program for a six-axis robot obstacle avoidance path planning method based on the Adam optimization algorithm. When the program for the six-axis robot obstacle avoidance path planning method based on the Adam optimization algorithm is executed by a processor, the steps of a six-axis robot obstacle avoidance path planning method based on the Adam optimization algorithm as described in any one of the above are implemented.

[0113] A six-axis robot obstacle avoidance path planning method and system disclosed in the present invention utilize the Adam algorithm to perform gradient descent optimization on the running path. Even in the face of a complex application environment, it is possible to achieve an obstacle avoidance effect in the path planning of the robot, thereby improving the running efficiency of the robot and reducing the collision rate on the traveling path.

[0114] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be electrical, mechanical, or other forms.

[0115] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units; they may be located in one place or distributed to multiple network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0116] In addition, in each embodiment of the present invention, the various functional units can all be integrated in one processing unit, or each unit can be separately used as one unit, or two or more units can be integrated in one unit; the above integrated units can be implemented in the form of hardware, or in the form of a combination of hardware and software functional units.

[0117] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed 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 including those of the above method embodiments. The aforementioned storage medium includes various media that can store program codes, such as removable storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0118] Alternatively, if the above integrated unit of the present invention is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the embodiments of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media that can store program codes, such as removable storage devices, ROM, RAM, magnetic disks, or optical discs.

Claims

1. A six-axis robot obstacle avoidance path planning method based on the Adam optimization algorithm, characterized in that: The following steps are involved: Obtain the path endpoint corresponding to the current six-axis robot, wherein the path endpoint includes a start point joint value and an end point joint value; Uniformly interpolating between the path endpoints to obtain path key points; Sampling is performed based on a preset step length, and the number of collision voxels is obtained by combining the path endpoint and the path key point, wherein specifically including sampling based on a preset step length, and obtaining a loss value based on the number of collision voxels, wherein the loss value The calculation formula is as follows: ； in, It is expressed as the number of samples taken between every two key path points. Represented as solving joint and joints The joint space distance, Is to find the joint value of the six-axis robot The number of collision voxels in the collision area with the environment, is the key point of the path, is the number of key points on the path, is the joint value of the six-axis robot; The loss function value is obtained based on the number of collision voxels, and the loss function is iteratively solved using the Adam optimization algorithm to obtain the planned path when the loss is zero.

2. According to claim 1, a six-axis robot obstacle avoidance path planning method based on the Adam optimization algorithm is characterized in that: The step of obtaining the path endpoint corresponding to the current six-axis robot specifically includes: The path endpoint is obtained based on the current working coordinate system of the six-axis robot, where: The starting point joint value is obtained based on the current position coordinates of the robot ; The end point joint value is obtained based on the position coordinates of the robot's destination .

3. A six-axis robot obstacle avoidance path planning method based on the Adam optimization algorithm according to claim 2, characterized in that: The uniform interpolation between the path endpoints to obtain the path key points specifically includes: Performing N uniform interpolations between the starting joint value and the end joint value to obtain the path key point, wherein the path key point is the joint value of the robot executing the target instruction; The calculation formula for obtaining N path key points is as follows: ； in, For the The joint values ​​corresponding to the key points of the path, is the interpolation function, is the number of key points on the path.

4. The six-axis robot obstacle avoidance path planning method based on the Adam optimization algorithm according to claim 3 is characterized in that: The interpolation function The calculation expression is as follows: ,in, is the actual calculated parameter value.

5. The six-axis robot obstacle avoidance path planning method based on the Adam optimization algorithm according to claim 4 is characterized in that: The method includes deleting redundant points on the planned path, wherein if the first When the loss is still zero when the path key point is reached, the current The key points of the path are deleted as the redundant points.

6. A six-axis robot obstacle avoidance path planning system based on the Adam optimization algorithm, characterized in that: The invention comprises a memory and a processor, wherein the memory comprises a six-axis robot obstacle avoidance path planning method program based on the Adam optimization algorithm, and the six-axis robot obstacle avoidance path planning method program based on the Adam optimization algorithm is executed by the processor to implement the following steps: Obtain the path endpoint corresponding to the current six-axis robot, wherein the path endpoint includes a start point joint value and an end point joint value; Uniformly interpolating between the path endpoints to obtain path key points; Sampling is performed based on a preset step length, and the number of collision voxels is obtained by combining the path endpoint and the path key point, wherein specifically including sampling based on a preset step length, and obtaining a loss value based on the number of collision voxels, wherein the loss value The calculation formula is as follows: ； in, It is expressed as the number of samples taken between every two key path points. Represented as solving joint and joints The joint space distance, Is to find the joint value of the six-axis robot The number of collision voxels in the collision area with the environment, is the key point of the path, is the number of key points on the path, is the joint value of the six-axis robot; The loss function value is obtained based on the number of collision voxels, and the loss function is iteratively solved using the Adam optimization algorithm to obtain the planned path when the loss is zero.

7. The six-axis robot obstacle avoidance path planning system based on the adam optimization algorithm according to claim 6 is characterized in that: The step of obtaining the path endpoint corresponding to the current six-axis robot specifically includes: The path endpoint is obtained based on the current working coordinate system of the six-axis robot, where: The starting point joint value is obtained based on the current position coordinates of the robot ; The end point joint value is obtained based on the position coordinates of the robot's destination .

8. The six-axis robot obstacle avoidance path planning system based on the Adam optimization algorithm according to claim 7 is characterized in that: The uniform interpolation between the path endpoints to obtain the path key points specifically includes: Performing N uniform interpolations between the starting joint value and the end joint value to obtain the path key point, wherein the path key point is the joint value of the robot executing the target instruction; The calculation formula for obtaining N path key points is as follows: ； in, For the The joint values ​​corresponding to the key points of the path, is the interpolation function, is the number of key points on the path.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a six-axis robot obstacle avoidance path planning method program based on the Adam optimization algorithm. When the six-axis robot obstacle avoidance path planning method program based on the Adam optimization algorithm is executed by the processor, the steps of the six-axis robot obstacle avoidance path planning method based on the Adam optimization algorithm as described in any one of claims 1 to 5 are implemented.

Images