Non-standard welding-oriented robot efficient collision-free path planning method and system

Abstract

The invention discloses a non-standard weldment-oriented robot efficient collision-free path planning method and system, and belongs to the technical field of robot welding, and the method comprises the following steps: S1, obtaining three-dimensional model data of a non-standard weldment and an operation environment, and extracting welding seam point cloud data; s2, carrying out feature extraction and downsampling on the welding seam point cloud data to generate a group of key path points; s3, global optimization is conducted on the welding path through an improved genetic algorithm, candidate paths are generated, and the welding sequence of the multiple sections of welding seams and the welding gun postures at the key path points are optimized at the same time through the improved genetic algorithm; s4, evaluating the candidate paths generated by the improved genetic algorithm through a multi-target fitness function including path length, attitude stability and collision risk; and S5, outputting an optimal collision-free welding path according to an evaluation result, and converting the optimal collision-free welding path into a smooth track which can be executed by the robot. According to the invention, the high-efficiency, high-quality and collision-free welding gun motion trail can be automatically generated.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of robot welding, and particularly relates to a robot efficiency collision-free path planning method and system for non-standard welding parts. BACKGROUND

[0002] Modern manufacturing is moving towards small batch, multi-variety and customization, which makes the demand for welding of non-standard structural parts grow continuously. Traditional robot welding programming mainly relies on two ways of "teach and reproduce" and "offline programming". Teach and reproduce requires experienced technicians to manually guide the robot to record path points. In the face of complex welds, this method not only consumes time and effort, but also is difficult to ensure accuracy. Although offline programming transfers the programming work from the production line to the computer, the data interaction between CAD and offline programming software often has a fault, and a large amount of manual intervention is needed to reset the welding parameters and path, resulting in a long production preparation period and poor flexibility.

[0003] Existing research shows that these traditional methods often rely on pre-defined rules and templates when dealing with unstructured environments and irregular workpieces, and have obvious lack of flexibility, making it difficult to adapt to changes in weld position and shape, thereby affecting the consistency and reliability of welding quality. In addition, for automated path planning, the academic community has proposed a variety of algorithms, such as the rapidly-exploring random tree (RRT) and its variants (RRT*) based on sampling, and the A* algorithm based on graph search. However, these algorithms often face challenges when applied to welding scenarios: they usually focus on finding the shortest collision-free path in geometry, but ignore the special requirements of welding process on path quality, such as the smoothness of welding gun pose, the continuity of angular velocity, and the dynamic adjustment of welding speed. Simply optimizing a single objective (such as path length) often reduces welding quality or fails to meet process constraints. Therefore, developing a multi-objective planning algorithm that can synergistically optimize efficiency, quality and safety has become a technical problem that needs to be solved in this field. SUMMARY

[0004] The purpose of the present application is to meet the actual needs and provide a robot efficiency collision-free path planning method and system for non-standard welding parts, which combines improved genetic algorithm and the collaborative planning framework of mainstream robot motion planning library, and can automatically generate efficient, high-quality and collision-free welding gun motion trajectories for non-standard workpieces with irregular geometric shapes and complex spatial welds.

[0005] The first purpose of the present application is to provide a robot efficiency collision-free path planning method for non-standard welding parts, comprising:

[0006] S1, obtaining three-dimensional model data of non-standard welding parts and working environment, and extracting weld point cloud data;

[0007] S2, feature extraction and down-sampling are performed on the weld seam point cloud data to generate a set of key path points;

[0008] S3, a global optimization is performed on the welding path by using an improved genetic algorithm to generate a candidate path, and the improved genetic algorithm simultaneously optimizes the welding sequence of the multi-segment weld seam and the welding gun posture at each key path point;

[0009] S4, the candidate path generated by the improved genetic algorithm is evaluated by using a multi-objective fitness function including path length, posture stability and collision risk;

[0010] S5, according to the evaluation result, an optimal collision-free welding path is output, and the optimal collision-free welding path is converted into a smooth trajectory executable by a robot.

[0011] Preferably, the feature extraction and down-sampling includes an adaptive method based on local geometric curvature of the weld seam, sparse sampling is performed in a flat area, and dense sampling is performed in a corner area.

[0012] Preferably, the improved genetic algorithm uses a hybrid coding chromosome, and the hybrid coding chromosome includes an integer arrangement part for representing the welding sequence of the weld seam and a real number or quaternion array part for representing the welding gun posture at each key point.

[0013] Preferably, in the initialization stage of the improved genetic algorithm, a sampling-based motion planning library is called to generate an initial collision-free path for part or all of the weld seam segments, and the initial collision-free path is injected into the initial population as a high-quality gene to accelerate the convergence of the algorithm.

[0014] Preferably, the evaluation is based on an Octomap octree map to model the environment and perform collision detection on the candidate path.

[0015] A second object of the present application is to provide a robot efficiency collision-free path planning system for non-standard welding parts, comprising:

[0016] A data acquisition module acquires three-dimensional model data of the non-standard welding part and the working environment, and extracts weld seam point cloud data;

[0017] A data preprocessing module performs feature extraction and down-sampling on the weld seam point cloud data to generate a set of key path points;

[0018] An optimization module performs global optimization on the welding path by using an improved genetic algorithm to generate a candidate path, and the improved genetic algorithm simultaneously optimizes the welding sequence of the multi-segment weld seam and the welding gun posture at each key path point;

[0019] an evaluation module for evaluating the candidate paths generated by the improved genetic algorithm through a multi-objective fitness function including path length, pose stability and collision risk;

[0020] an output module for outputting the optimal collision-free welding path according to the evaluation result and converting the optimal collision-free welding path into a smooth trajectory executable by the robot.

[0021] Preferably, a process coupling optimization model is further included for adaptively adjusting the planned welding speed according to the local curvature of the path.

[0022] Preferably, the feature extraction and down-sampling include an adaptive method based on the local geometric curvature of the weld to perform sparse sampling in flat areas and dense sampling in corner areas.

[0023] The improved genetic algorithm adopts a hybrid coding chromosome including an integer permutation part for representing the welding sequence of the weld and a real number or quaternion array part for representing the pose of the welding gun at each key point.

[0024] In the initialization stage of the improved genetic algorithm, a sampled motion planning library is called to generate initial collision-free paths for part or all of the weld segments, which are injected into the initial population as high-quality genes to accelerate the convergence of the algorithm.

[0025] The evaluation is based on an Octomap octree map for modeling the environment and performing collision detection on the candidate paths.

[0026] A third object of the present application is to provide a computer-readable storage medium storing a computer program which, when executed by a processor, implements the robot collision-free path planning method for non-standard workpieces as described above.

[0027] A fourth object of the present application is to provide a computer program product comprising a computer program which, when executed by a processor, implements the robot collision-free path planning method for non-standard workpieces as described above.

[0028] The present application has the advantages and positive effects that:

[0029] 1) Efficiency improvement: The present invention exhibits significant superiority in both "planning time-consuming" and "path length", two key performance indicators. Specifically, compared with the traditional RRT algorithm, the present invention can significantly reduce the time required for planning on average and effectively shorten the length of the path by introducing advanced algorithm mechanisms. This significant efficiency improvement is mainly due to the global optimization capability of genetic algorithm (GA) and the synergistic effect of open source motion planning library (OMPL), which makes the algorithm quickly converge to the optimal solution in the search process, thereby greatly improving the overall planning efficiency.

[0030] 2) Quality improvement: In terms of pose smoothness, the present invention ensures smooth and continuous changes in the pose of the robot arm during welding through refined algorithm design and optimization, thereby effectively ensuring the stability and reliability of the welding process. In addition, the present invention also focuses on avoiding collision risks during welding, significantly improving the safety of the production process and reducing equipment damage and production accidents caused by collisions, further improving the overall production quality. BRIEF DESCRIPTION OF DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0032] Figure 1 is a flowchart of the preferred embodiment of the present invention;

[0033] Figure 2 is another flowchart of the preferred embodiment of the present invention;

[0034] Figure 3 is a schematic diagram of the outrigger model in the preferred embodiment of the present invention. DETAILED DESCRIPTION

[0035] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0036] Please refer to Figure 1 and Figure 2

[0037] A robot efficient collision-free path planning method for non-standard welding parts, comprising the following steps:

[0038] S1, acquire three-dimensional model data of non-standard welding parts and working environment, and extract welding seam point cloud data;

[0039] The purpose of this step is to acquire three-dimensional model data of non-standard welding parts and working environment, to comprehensively capture the structural features of the welding parts and detailed information of the working environment through high-precision scanning technology, and on this basis, to further extract point cloud data of the welding seam area, ensuring the accuracy and integrity of the point cloud data, and providing reliable data support for subsequent welding process analysis and optimization.

[0040] S2, feature extraction and down-sampling of the welding seam point cloud data to generate a set of key path points;

[0041] The purpose of this step is to perform a detailed and accurate feature extraction process on the welding seam point cloud data, to effectively identify the key features of the welding seam by using advanced algorithms and technical means to analyze the geometric characteristics and structural information in the point cloud data. On this basis, further efficient down-sampling processing of the extracted features is performed, aiming to reduce data redundancy and optimize data processing efficiency, and finally a set of refined and representative key path points are generated. These key path points not only accurately reflect the overall shape and local details of the welding seam, but also provide important data support for subsequent welding process optimization and quality control.

[0042] S3, global optimization of the welding path using an improved genetic algorithm to generate candidate paths, the improved genetic algorithm simultaneously optimizing the welding sequence of multiple welding seams and the welding gun attitude at each key path point;

[0043] In the global optimization process of the welding path, this embodiment uses an improved genetic algorithm, which can generate a series of candidate welding paths. These candidate paths not only consider the overall optimization of the welding sequence, but also make detailed adjustments to the welding sequence of multiple welding seams. In addition, the improved genetic algorithm can also optimize the welding gun attitude at each key path point, ensuring that the welding gun can operate at the most appropriate angle and position during welding. Through such optimization, a welding path that is both efficient and of high quality can be obtained, thereby improving welding efficiency and welding quality.

[0044] S4, evaluate the candidate paths generated by the improved genetic algorithm through a multi-objective fitness function containing path length, attitude stability and collision risk;

[0045] The candidate paths generated by the improved genetic algorithm are comprehensively and meticulously evaluated through a multi-objective fitness function that considers multiple key factors such as path length, posture stability, and collision risk. This fitness function aims to quantitatively analyze the pros and cons of the paths from multiple dimensions, ensuring that the selected path is not only optimal in length but also performs well in maintaining posture and avoiding collisions, thereby providing more reliable and efficient decision-making basis for path planning.

[0046] S5、According to the evaluation results, output the optimal collision-free welding path, and convert the optimal collision-free welding path into a smooth trajectory executable by the robot.

[0047] In order to better understand the technical solutions of the present application, the following is a non-limiting explanation:

[0048] The feature extraction and down-sampling is an adaptive method based on the local geometric curvature of the weld, which performs sparse sampling in flat areas with low curvature and dense sampling in corner areas with high curvature.

[0049] The feature extraction and down-sampling method is an adaptive technology that intelligently adjusts based on the local geometric curvature of the weld. Specifically, when the local area of the weld exhibits low curvature, i.e., relatively flat parts, the method will adopt a sparse sampling strategy. This means that in these flat areas, the density of feature extraction and down-sampling will be relatively low, as the features in flat areas do not change significantly and do not require too many sampling points to capture details. Conversely, when the local area of the weld has high curvature, i.e., there are corners or other complex shapes, the method will automatically adjust to a dense sampling strategy. In the corner area with high curvature, the density of feature extraction and down-sampling will increase to ensure that more detailed information can be captured, thus more accurately describing the features of the weld. This adaptive feature extraction and down-sampling method can intelligently adjust the sampling strategy according to the actual shape and structure of the weld, ensuring both the accuracy of feature extraction and the efficiency of sampling, and is a very intelligent and efficient technology.

[0050] The improved genetic algorithm uses a hybrid encoding chromosome, which includes an integer permutation part for representing the welding sequence of the weld and a real number or quaternion array part for representing the posture of the welding gun at each key point.

[0051] The improved genetic algorithm in this embodiment adopts a hybrid coding chromosome, which is composed of two main parts. The first part is an integer arrangement part for representing the welding sequence of the weld, which accurately describes the welding sequence of the weld through the arrangement order of integers, ensuring the rationality and efficiency of the welding process. The second part is a real number or quaternion array part for representing the pose of the welding gun at each key point, which records the specific pose information of the welding gun at each key point in detail through the array form of real numbers or quaternions, including angle, direction and other key parameters, so as to ensure that the welding gun can operate accurately according to the preset pose during the welding process, improving the welding quality and precision. Through this hybrid coding method, the improved genetic algorithm can more comprehensively and finely handle complex problems in the welding process, optimize the welding sequence and welding gun pose, and further improve the overall performance and effect of the welding process.

[0052] The improved genetic algorithm adopts an adaptive mutation strategy related to the curvature of the weld, so that the pose genes corresponding to the positions with higher curvature on the path have a higher mutation probability.

[0053] In the initialization stage of the improved genetic algorithm, a high-quality initial collision-free path is generated for part or all of the weld segments by calling a motion planning library based on sampling (such as OMPL), and is injected into the initial population as a high-quality gene to accelerate the convergence of the algorithm.

[0054] The evaluation of collision risk is based on the Octomap octree map to model the environment and perform human-effective collision detection on the candidate path.

[0055] It also includes a process coupling optimization that adaptively adjusts the planned welding speed according to the local curvature of the path, reducing the speed at positions with larger curvature to ensure welding quality.

[0056] A robot efficient collision-free path planning system for non-standard welding parts, comprising:

[0057] A data acquisition module acquires three-dimensional model data of the non-standard welding part and the working environment, and extracts weld point cloud data;

[0058] A data preprocessing module extracts features and down-samples the weld point cloud data to generate a set of key path points;

[0059] An optimization module uses an improved genetic algorithm to globally optimize the welding path and generate a candidate path, the improved genetic algorithm simultaneously optimizes the welding sequence of multiple welds and the welding gun pose at each key path point;

[0060] An evaluation module evaluates the candidate path generated by the improved genetic algorithm through a multi-objective fitness function including path length, pose stability and collision risk.

[0061] an output module, configured to output the optimal collision-free welding path according to the evaluation result, and convert the optimal collision-free welding path into a smooth trajectory executable by the robot.

[0062] In one of the specific embodiments, a process coupling optimization model is further included, which is configured to adaptively adjust the planned welding speed according to the local curvature of the path.

[0063] In one of the specific embodiments, the feature extraction and down-sampling includes an adaptive method based on the local geometric curvature of the weld, which performs sparse sampling in flat areas and dense sampling in corner areas.

[0064] In one of the specific embodiments, the improved genetic algorithm adopts a hybrid encoding chromosome, which includes an integer permutation part for representing the welding sequence of the weld and a real number or quaternion array part for representing the pose of the welding gun at each key point.

[0065] In one of the specific embodiments, in the initialization stage of the improved genetic algorithm, a sampling-based motion planning library is called to generate initial collision-free paths for part or all of the weld segments, which are injected into the initial population as high-quality genes to accelerate the convergence of the algorithm.

[0066] In one of the specific embodiments, the evaluation is based on an Octomap octree map to model the environment and perform collision detection on the candidate paths.

[0067] The specific implementation process of the above system includes:

[0068] Step 1: Environment and task initialization: the system first loads the workpiece model from CAD, the weld point cloud data, and the STL model of the surrounding equipment. An initial three-dimensional environment containing all static obstacles is constructed in the MoveIt planning scene. The weld point cloud is parsed into a series of target path segments to be welded.

[0069] Step 2: Weld feature extraction: the original weld point cloud is called to perform adaptive down-sampling using a feature extraction algorithm, generating a set of sparse but accurate key path points that can accurately describe the weld geometry.

[0070] Step 3: Hybrid initialization population generation: to accelerate the convergence of the genetic algorithm, a hybrid initialization strategy is adopted. By calling the OMPL planning library of MoveIt, high-quality collision-free initial paths for the weld segments are quickly generated and decoded into chromosome form, which are injected into the initial population.

[0071] Step 4: Genetic Algorithm Iterative Optimization: The algorithm continuously generates new solutions (chromosomes) through selection, crossover, and mutation operations. Each solution is evaluated by a comprehensive fitness function that quantifies the path's length, pose smoothness, and collision risk.

[0072] Step 5: Path Generation and Validation: After the genetic algorithm iterations end, the chromosome with the highest fitness is output. This optimal solution (containing the weld sequence and weld gun poses for each key point) is converted into a smooth trajectory. This trajectory will undergo final, more detailed collision validation in MoveIt and be visualized.

[0073] Step 6: Output and Robot Execution: After validation, standard robot trajectory instructions (such as JointTrajectory messages) are generated and published to the ROS network, which are subscribed to by the robot driver and control the robot to execute accurately.

[0074] Non-standard weld feature extraction and path point downsampling:

[0075] Adaptive downsampling based on local geometric features (especially curvature), including:

[0076] For straight segments (local curvature below threshold ε1), perform significant uniform sparse sampling, retaining only the start and end key points that represent the direction of the straight segment.

[0077] For high-curvature corner segments (local curvature greater than threshold ε2), retain all original points or perform slight encryption sampling to ensure that the path accurately fits the sharp geometric changes of the weld.

[0078] For ease curve segments (curvature between ε1 and ε2), perform proportional sampling based on curvature.

[0079] The improved genetic algorithm (GA) hybrid initialization population generation method includes two parts of chromosome encoding: optimized welding sequence and weld gun pose.

[0080] 1) Weld sequence encoding (integer permutation). Assuming there are N independent weld segments on the workpiece, this part is a permutation combination of integers (such as 1 to N) with a length of N.

[0081] 2) Key point pose encoding (real number / quaternion array). This part is a variable-length real number array that stores the weld gun poses of all weld key points. The pose is usually represented by a quaternion q(w, x, y, z), and the initial pose is determined by the normal vector information of the weld point cloud, and the genetic algorithm will fine-tune and optimize it.

[0082] This embodiment proposes an adaptive mutation strategy related to the curvature of the weld seam. In the flat weld seam segment, the mutation probability should be lower to maintain the excellent genes; at the corner with large curvature, use more mutation probability to encourage the algorithm to explore more possibilities.

[0083] The mutation probability P_mutation(i) of the i-th pose gene is determined by the following formula:

[0084] P_mutation(i)=P_base+k*Curvature(i), where P_base is the base mutation rate, k is the coefficient, and Curvature(i) is the normalized curvature.

[0085] The improved genetic algorithm (GA) fitness function (Fitness Function) is proposed in this invention, a maximum multi-objective fitness function F_total is proposed, which is as follows:

[0086] Fitness function F_total=w1*F_length+w2*F_pose+w3*F_collision

[0087] Where w1, w2, w3 are weight coefficients, used to balance the importance of different optimization objectives. The definition of each term is as follows:

[0088] F_length (path length): calculate the total distance of the welding torch TCP (tool center point) traveled, including the length of all welding paths and the transition path length between the weld seams.

[0089] F_pose (pose stability): measures the smoothness of the welding torch pose throughout the welding process.

[0090] F_collision (collision cost): when evaluating each chromosome, a collision detection module is called to check the generated path. If any part of the path collides with the environment, F_collision will be assigned a very large penalty value, ensuring that any solution with collision risk is eliminated in the selection process.

[0091] Please refer to Figure 3 , the workpiece leg is the welding object, and the welding equipment is a 6-axis welding robot;

[0092] Step 1: Environment and task initialization. The system first loads the workpiece leg CAD model and the surrounding environment STL model as shown in Figure 2 At the same time, the pre-extracted weld seam point cloud data is imported, and its data structure is as follows: [

[0094] {weld seam identifier: c169, weld seam length: 4.0 mm},

[0095] {weld id: f698, weld length: 63.0 mm},

[0096] {weld id: d3fe, weld length: 100.0 mm} ]

[0098] Data is parsed in the MoveIt! planning scene, building an initial 3D environment containing static obstacles.

[0099] Step 2: Weld feature extraction and adaptive down-sampling. Perform adaptive down-sampling based on local geometric curvature for the original weld point cloud. In this example, set the low curvature threshold ε1 = 0.1 and the high curvature threshold ε2 = 0.8 (after normalization).

[0100] For long straight welds on the workpiece legs, whose curvature is lower than ε1, the algorithm performs sparse sampling, retaining only the first and last key points;

[0101] For L-shaped corner welds, whose curvature is higher than ε2, the algorithm retains most of the original points to accurately describe the geometric changes, thus generating a set of sparse but key path points.

[0102] Step 3: Improved genetic algorithm optimization. Use an improved genetic algorithm (GA) to globally optimize the welding sequence and welding gun pose. The core parameters of the algorithm are set as follows:

[0103] Hybrid initialization: Call the OMPL planning library to quickly generate high-quality collision-free initial paths for each weld segment, and after decoding, inject them as high-quality genes into the initial population to accelerate convergence.

[0104] Adaptive mutation strategy: Use an adaptive mutation strategy related to weld curvature.

[0105] The mutation probability of the i-th pose gene is determined by the following formula:

[0106] Pmutation(i) = Pbase + k * Curvature(i)

[0107] In this example, set the base mutation rate P_base = 0.01 and the curvature influence coefficient k = 0.05. This allows the welding gun pose at corners (high curvature) to have more opportunities for exploratory mutation, while maintaining stability at flat sections (low curvature).

[0108] Multi-objective fitness function: Use the weighted sum method to evaluate the pros and cons of each candidate solution (chromosome). The fitness function is as follows:

[0109] Ftotal = w1 * Flength + w2 * Fpose + w3 * Fcollision

[0110] where F length is the total path length, F pose evaluates the smoothness of the pose, and F collision is the collision cost. In this embodiment, to balance the efficiency, quality, and safety, the weight coefficients are set as w1=0.4, w2=0.3, and w3=0.3. When a collision is detected, F collision is assigned a very large penalty value, ensuring that the solution is eliminated in evolution.

[0111] Step 4: Path generation and output

[0112] After a set number of iterations, the genetic algorithm outputs the chromosome with the highest fitness. This optimal solution (containing the optimal welding sequence and pose of the welding gun at each key point) is converted into a smooth robot trajectory and subjected to a final collision check in MoveIt!. After passing the check, standard robot control instructions (such as ROS JointTrajectory messages) are generated and sent to the robot for execution.

[0113] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the robot collision-free path planning method for non-standard welding pieces described above.

[0114] A computer program product comprising a computer program, which, when executed by a processor, implements the robot collision-free path planning method for non-standard welding pieces described above.

[0115] In the above embodiments, all or part can be implemented by software, hardware, firmware, or any combination thereof. When implemented in whole or in part in the form of a computer program product, the computer program product comprises one or more computer instructions. When the computer program instructions are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another, for example, the computer instructions can be transferred from one website, computer, server, or data center to another via wired (such as coaxial cable, optical fiber, digital subscriber line (DSL), or wireless (such as infrared, wireless, microwave, etc.)) mode. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media sets. The available media can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid-state disk (SSD)), etc.

[0116] The above merely describes the preferred embodiments of the present application, and it should be pointed out that for those skilled in the art, several improvements and refinements can be made without departing from the principles of the present application, and these improvements and refinements should also be considered as falling within the scope of protection of the present application.

Claims

1. A method for efficient collision-free path planning of robots for non-standard welded parts, characterized in that, include: S1. Obtain 3D model data of non-standard welded parts and working environment, and extract weld point cloud data; S2. Perform feature extraction and downsampling on the weld seam point cloud data to generate a set of critical path points; S3. An improved genetic algorithm is used to globally optimize the welding path and generate candidate paths. The improved genetic algorithm simultaneously optimizes the welding sequence of multiple weld seams and the welding torch posture at each critical path point. S4. The candidate paths generated by the improved genetic algorithm are evaluated using a multi-objective fitness function that includes path length, attitude stability, and collision risk. S5. Based on the evaluation results, output the optimal collision-free welding path and convert the optimal collision-free welding path into a smooth trajectory that can be executed by the robot.

2. The robot human-friendly collision-free path planning method for non-standard welded parts according to claim 1, characterized in that, The feature extraction and downsampling include: an adaptive method based on the local geometric curvature of the weld, which performs sparse sampling in straight areas and dense sampling in corner areas.

3. The robot collision-free path planning method for non-standard welded parts according to claim 1, characterized in that, The improved genetic algorithm employs a hybrid encoded chromosome, which includes an integer arrangement portion representing the welding sequence of the weld seam and a real number or quaternion array portion representing the welding torch posture at each key point.

4. The robot collision-free path planning method for non-standard welded parts according to claim 1, characterized in that, In the initialization phase of the improved genetic algorithm, a sampling-based motion planning library is invoked to generate initial collision-free paths for some or all weld segments, and these paths are injected into the initial population as high-quality genes to accelerate algorithm convergence.

5. The robot collision-free path planning method for non-standard welded parts according to claim 1, characterized in that, The evaluation is based on Octomap octree map modeling of the environment and collision detection of candidate paths.

6. A robot-based human-like collision-free path planning system for non-standard welded parts, characterized in that, include: The data acquisition module acquires 3D model data of non-standard welded parts and the working environment, and extracts weld point cloud data; The data preprocessing module performs feature extraction and downsampling on the weld seam point cloud data to generate a set of critical path points; The optimization module uses an improved genetic algorithm to globally optimize the welding path and generate candidate paths. The improved genetic algorithm simultaneously optimizes the welding sequence of multiple weld segments and the welding torch posture at each critical path point. The evaluation module evaluates the candidate paths generated by the improved genetic algorithm using a multi-objective fitness function that includes path length, attitude stability, and collision risk. The output module outputs the optimal collision-free welding path based on the evaluation results and converts the optimal collision-free welding path into a smooth trajectory that can be executed by the robot.

7. The robot-based human-efficiency collision-free path planning system for non-standard welded parts according to claim 6, characterized in that, It also includes a process coupling optimization model that adaptively adjusts the planned welding speed based on the local curvature of the path.

8. The robot human-efficiency collision-free path planning system for non-standard welded parts according to claim 6, characterized in that: The feature extraction and downsampling include: an adaptive method based on the local geometric curvature of the weld, which performs sparse sampling in straight areas and dense sampling in corner areas; The improved genetic algorithm uses a hybrid encoded chromosome, which includes an integer arrangement part for representing the welding sequence of the weld seam and a real number or quaternion array part for representing the welding gun posture of each key point. In the initialization phase of the improved genetic algorithm, a sampling-based motion planning library is called to generate initial collision-free paths for some or all weld segments, and these paths are injected into the initial population as high-quality genes to accelerate algorithm convergence. The evaluation is based on Octomap octree map modeling of the environment and collision detection of candidate paths.

9. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the robot human-like collision-free path planning method for non-standard welded parts as described in any one of claims 1-4.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the efficient collision-free path planning method for robots targeting non-standard welded parts as described in any one of claims 1-4.

Images