Multi-axis robot path planning optimization method and system

Abstract

The invention relates to the technical field of path optimization, in particular to a multi-axis robot path planning optimization method and system, and the method comprises the following steps: collecting the current position and the start-stop interval of a multi-motion-axis joint, calculating the motion distance of each joint, converting the motion distance into an interval occupation ratio, and generating an initial path point through the interpolation of the current position; the method comprises the following steps of: selecting a plurality of path point joint combinations, performing occupation proportion comparison on the plurality of path point joint combinations to determine a dominant constraint direction, performing mapping cutting on other joint motion ranges according to a residual interval proportion, screening path points meeting constraint conditions, and verifying path sequence consistency and joint displacement continuity to form an optimized motion path of the robot. Through multi-motion-axis joint interval proportional description, an interval occupation state is introduced to participate in path screening, a dominant constraint direction is determined, and other joint intervals are mapped and cut, so that multi-axis motion proportions are coordinated, invalid displacement is reduced, and collaboration and motion smoothness are improved.

Description

Technical Field

[0001] This invention relates to the field of path optimization technology, and in particular to a method and system for path planning and optimization of multi-axis robots. Background Technology

[0002] The field of path optimization technology mainly involves technologies related to the planning, calculation, and adjustment of motion trajectories in robotics and automation systems. Its core aspects include the establishment of multi-axis motion models, the representation of spatial trajectories, the motion mapping between joints and end effectors, the generation of discrete points on the trajectory, and the handling of constraints in the motion process. This technical field usually revolves around the motion path of a robot from its starting position to its target position within the workspace. By systematically designing path nodes, the sequence of joint angle changes, the motion timing, and constraints, it supports the motion planning needs of industrial robots in scenarios such as assembly, handling, and welding.

[0003] One traditional method and system for multi-axis robot path planning optimization refers to establishing a kinematic model of a multi-axis robot, discretizing the workspace based on the correspondence between joint angle variables and end-effector pose, generating several candidate path points, and combining and adjusting the order of path points according to the range of motion, speed limits, and continuity requirements of each joint. It often uses a segmented interpolation method to calculate the joint motion trajectory between adjacent path points, and plans and adjusts the generated multi-axis motion path by calculating the path length, joint angle change, or time parameters to determine the overall motion path of the robot.

[0004] Current multi-axis robot path planning mainly relies on workspace discretization and segmented interpolation. Path point generation focuses on geometric continuity and joint range verification, lacking an overall consideration of the occupancy relationship of multiple motion axis intervals. In multi-axis cooperative motion scenarios, it is prone to joint load distribution imbalance. Path point combination depends on predetermined order rules, making it difficult to reflect the primary and secondary influence of different joints in the whole motion. Constraint processing is mostly concentrated on single-axis limits or local continuity, resulting in coarse path selection granularity. Redundant displacements may occur during the motion, and local joint strokes are frequently compressed or amplified, affecting the overall motion smoothness and execution stability. Summary of the Invention

[0005] To address the technical problems existing in the prior art, this invention provides a multi-axis robot path planning and optimization method, comprising the following steps: S1: Collect the current position of the joints corresponding to the multiple motion axes of the dual-sided synchronous labeling multi-axis robot, the start and end positions of the motion, calculate the joint motion interval distance and the total motion interval length and convert them into the interval occupancy ratio, perform linear interpolation with the start and end positions of the motion to generate the initial path point set; S2: Obtain the joint position group corresponding to the multi-path points in the initial path point set, compare the interval occupancy ratio of the motion axes in the multi-path points with the interval occupancy status, select the motion axis with the optimal ratio, and generate dominant joint constraint data; S3: Based on the dominant joint constraint data, obtain the proportion of the remaining joint motion range of the corresponding motion axis, extract the total joint motion range length of the corresponding joint, perform boundary conversion according to the proportion of the remaining joint motion range, and generate the path joint constraint range. S4: Based on the joint position combination corresponding to the multi-path points in the initial path point set, determine whether it is within the path joint constraint range and filter the path points to generate an optimized path point set; S5: Based on the optimized path point set, obtain the joint positions corresponding to the multiple path points and perform path point sequence consistency verification and joint displacement continuity judgment to generate an optimized robot motion path.

[0006] As a further embodiment of the present invention, the initial path point set includes a path point index sequence, multi-motion axis interpolated joint positions, and interval occupancy status; the dominant joint constraint data includes the dominant motion axis identifier, the corresponding joint interval occupancy ratio, and constraint priority parameters; the path joint constraint range includes the upper and lower bounds of motion axis joint clipping, the remaining motion interval ratio mapping, and the set of feasible joint intervals; the optimized path point set includes path points within constraints and combinations of joint positions; and the optimized robot motion path includes the joint position temporal trajectory, the path point sequence, and the continuous change sequence of joint displacement.

[0007] As a further aspect of the present invention, the specific steps of S1 are as follows: S101: Collect the current position, start and end position of the joints corresponding to the multi-axis motion of the double-sided synchronous labeling multi-axis robot, calculate the total joint motion range length, perform difference calculation on the current position and start and end position of the multi-axis joints, perform normalization processing on the multi-axis span based on the maximum stroke range of the multi-axis, and generate a set of joint motion range parameters. S102: Based on the joint motion interval parameter set, perform a proportional mapping operation according to the multi-axis normalized interval length and the corresponding stroke span, determine the state based on the mapping result and the interval start and end markers, and serialize and rearrange the occupancy markers according to the motion axis order to obtain the interval occupancy state sequence. S103: Call the interval occupancy state sequence, perform linear interpolation operation according to the multi-axis occupancy interval ratio and the corresponding joint current position value, align and reorganize the interpolation results according to time, and associate the path order of the joint position arrangement to generate an initial path point set.

[0008] As a further aspect of the present invention, the specific steps of S2 are as follows: S201: Obtain the joint position combination corresponding to the multi-path points in the initial path point set, perform combination mapping on the multi-motion axis joint position values ​​for the multi-path point index, and vectorize and rearrange the multi-motion axis joint positions under the same temporal index to generate a path point joint combination sequence. S202: Based on the path point joint combination sequence and the interval occupancy status sequence, perform a comparison of the interval occupancy ratio corresponding to multiple motion axes within multiple path points, and judge and sort the difference between the multi-axis occupancy ratio under the same path point and the preset ratio benchmark value to obtain a motion axis occupancy ratio comparison identifier set. S203: Based on the set of motion axis occupancy ratio comparison identifiers, perform ratio filtering and judgment for multi-path points, select the motion axis with the optimal occupancy ratio sorting position, and perform constraint mapping between the selected motion axis index and the corresponding joint position to generate dominant joint constraint data.

[0009] As a further aspect of the present invention, the proportional benchmark value is determined by summarizing the interval occupancy ratios of multiple motion axes in the interval occupancy state sequence within the initial path point set, collecting the occupancy ratio values ​​of the same type of motion axis under the multi-path point index, and performing an average calculation on the value set.

[0010] As a further aspect of the present invention, the specific steps of S3 are as follows: S301: Obtain the proportion of the remaining joint motion interval for the corresponding motion axis based on the dominant joint constraint data, call the joint motion interval parameter for the dominant motion axis, perform proportion calculation on the interval length value, and arrange the calculation results according to the path point index to generate a sequence of remaining joint interval proportions. S302: Based on the remaining joint interval ratio sequence, obtain the joint motion interval parameters corresponding to the other motion axes in the path planning, use the remaining joint interval ratio sequence to perform boundary conversion on the total joint motion interval length of the multi-motion axis, calculate the effective interval boundary of the other motion axes under the current ratio, and obtain the reverse clipping interval boundary set; S303: Based on the reverse clipping interval boundary set, aggregate multi-motion axis clipping intervals for multi-path point index, perform joint constraint arrangement on the clipped intervals and the intervals corresponding to the dominant motion axis, and perform interval continuity verification on the arrangement results to generate path joint constraint range.

[0011] As a further aspect of the present invention, the specific steps of S4 are as follows: S401: Based on the joint position combination corresponding to the multi-path points in the initial path point set, collect the multi-joint position values ​​under the multi-path point index, align the joint position values ​​with the corresponding joint dimension index and perform structured aggregation to generate a path point joint position combination set. S402: Based on the set of path point joint positions and the range of path joint constraints, perform interval assignment judgment on each dimension of the joint position vector, mark the path point index that exceeds the interval boundary as invalid, and obtain the set of path points within the constraints. S403: Based on the set of path points within the constraints, perform a filtering and reorganization operation on the path points corresponding to the initial set of path points, and arrange the filtered path points in the original path order to generate an optimized set of path points.

[0012] As a further aspect of the present invention, the specific steps of S5 are as follows: S501: Based on the optimized path point set, obtain the multi-joint position values ​​corresponding to the multi-path points, and perform mapping between the joint number and the position value to form a joint position vector sequence and perform structured combination operation to generate a path point joint position sequence. S502: Based on the path point joint position sequence, perform a sequence consistency check on adjacent path points, determine the sequence arrangement state according to the path point index increment relationship, and remove index positions that do not conform to the increment relationship to obtain a sequence of joint positions with consistent order. S503: Based on the sequential joint position sequence, calculate the joint displacement continuity metric for adjacent joint position vectors, retain the path point sequence located at the joint displacement continuity threshold and perform path connection operation to generate an optimized robot motion path.

[0013] As a further aspect of the present invention, the joint displacement continuity threshold is determined by collecting the displacement differences of corresponding joint dimensions under adjacent path points to form a set of joint displacement differences, performing a sorting operation on the set of differences, and extracting the median value of the multi-joint displacement difference distribution.

[0014] A multi-axis robot path planning and optimization system, comprising: The path generation module collects the current joint position, start and end position of motion of the joints corresponding to the multi-axis motion axis of the double-sided synchronous labeling multi-axis robot, calculates the joint motion interval distance and the total motion interval length and converts them into the interval occupancy ratio, performs linear interpolation with the start and end positions of motion, generates an initial path point set and passes it to the main analysis module. The dominant analysis module obtains the joint position group corresponding to the multi-path points in the initial path point set, performs a comparison of the interval occupancy ratio of the motion axes within the multi-path points in combination with the interval occupancy status, selects the motion axis with the optimal ratio, generates dominant joint constraint data, and transmits it to the interval constraint module. The interval constraint module obtains the proportion of the remaining joint motion interval of the corresponding motion axis based on the dominant joint constraint data, extracts the total joint motion interval length of the corresponding joint, performs boundary conversion according to the proportion of the remaining joint motion interval, generates the path joint constraint range, and passes it to the path filtering module. The path filtering module determines whether the path points are within the path joint constraint range based on the joint position combination corresponding to the multi-path points in the initial path point set, filters the path points, generates an optimized path point set, and passes it to the motion optimization module. The motion optimization module, based on the optimized path point set, obtains the joint positions corresponding to multiple path points and performs path point sequence consistency verification and joint displacement continuity judgment to generate an optimized robot motion path.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by proportionalizing the distance between joint intervals of multiple motion axes, the occupancy status of the intervals is introduced to participate in the generation and screening of path points. This allows the path planning process to reflect the relative role of each joint in the overall stroke. The constraint direction of the dominant joint is determined by comparing the occupancy ratio, and the motion intervals of the remaining joints are mapped and trimmed accordingly. This enables the multi-axis motion range to converge in a coordinated manner under a unified proportional relationship. The path point screening and sequence verification are combined with continuity judgment to reduce the accumulation of invalid displacements. This allows the generated path to be improved simultaneously in terms of multi-axis coordination, joint load balancing, and motion smoothness. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the steps of the present invention; Figure 2 This is a detailed schematic diagram of S1 of the present invention; Figure 3 This is a detailed schematic diagram of S2 of the present invention; Figure 4 This is a detailed schematic diagram of S3 of the present invention; Figure 5 This is a detailed schematic diagram of S4 of the present invention; Figure 6 This is a detailed schematic diagram of S5 of the present invention; Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0018] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0019] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0020] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0021] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0022] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0023] Please see Figure 1 This invention provides a method for optimizing path planning for a multi-axis robot, comprising the following steps: S1: Collect the current position of the joints corresponding to the multiple motion axes of the dual-sided synchronous labeling multi-axis robot, the start and end positions of the motion, calculate the joint motion interval distance and the total motion interval length and convert them into the interval occupancy ratio, perform linear interpolation with the start and end positions of the motion to generate the initial path point set; S2: Obtain the joint position group corresponding to the multipath points in the initial path point set, and perform interval occupancy ratio comparison on the motion axis within the multipath points in combination with the interval occupancy status, and select the motion axis with the optimal ratio to generate dominant joint constraint data; S3: Based on the dominant joint constraint data, obtain the proportion of the remaining joint motion range of the corresponding motion axis, extract the total joint motion range length, perform boundary conversion according to the proportion of the remaining joint motion range, and generate the path joint constraint range. S4: Based on the combination of joint positions corresponding to multipath points in the initial path point set, determine whether they are within the path joint constraint range and filter path points to generate an optimized path point set; S5: Based on the optimized path point set, obtain the joint positions corresponding to multiple path points and perform path point sequence consistency verification and joint displacement continuity judgment to generate an optimized robot motion path.

[0024] The initial path point set includes a path point index sequence, multi-axis interpolated joint positions, and interval occupancy status. The dominant joint constraint data includes the dominant motion axis identifier, the corresponding joint interval occupancy ratio, and constraint priority parameters. The path joint constraint range includes the upper and lower bounds of motion axis joint clipping, the remaining motion interval ratio mapping, and the set of feasible joint intervals. The optimized path point set includes path points within constraints and combinations of joint positions. The optimized robot motion path includes the temporal trajectory of joint positions, the path point sequence, and the continuous change sequence of joint displacements.

[0025] Please see Figure 2 The specific steps of S1 are as follows: S101: Collect the current position, start and end position of the joints corresponding to the multi-axis motion of the double-sided synchronous labeling multi-axis robot, calculate the total joint motion range length, perform difference calculation on the current position and start and end position of the multi-axis joints, perform normalization processing on the multi-axis span based on the maximum stroke range of the multi-axis, and generate a set of joint motion range parameters. First, data acquisition is performed on a six-axis industrial robot, model ABBIRB1600, used for double-sided labeling of irregularly shaped workpieces. Using absolute encoders deployed at the robot joints, the controller reads the current joint position data of the six motion axes (defined as axes 1 to 6) in real time with a sampling period of 1 millisecond. Simultaneously, it retrieves the motion start position data and motion end position data pre-stored in the path planning controller's memory. These two sets of data correspond to a complete spatial movement task from the material picking point to the labeling point. After receiving the above three sets of data, the interval length is calculated independently for each motion axis. Specifically, the value of the motion end position is subtracted from the value of the motion start position, and the absolute value of the difference is taken to obtain the total joint motion interval length of each axis. Subsequently, a difference calculation is performed, subtracting the corresponding motion start position value from the current joint position value of each axis to obtain the current displacement of each axis. Based on this, a normalization process is performed on the multi-axis span. First, it is determined whether the total joint motion range length is 0. If the length is 0 (i.e., the starting position equals the ending position), the joint motion range parameter of that axis is directly set to 0 to avoid division by zero errors. If the length is not 0, the current displacement of each axis is divided by the corresponding total joint motion range length to obtain a dimensionless value between 0 and 1, which is the joint motion range parameter. To ensure the accuracy of data processing, the actual implementation data shown in Table 1 is used for explanation. The collected starting position of axis 1 is 10.00 degrees, the ending position is 50.00 degrees, and the current joint position is 15.00 degrees. The total joint motion range length of axis 1 is calculated as 50.00 - 10.00 = 40.00 degrees; the current displacement of axis 1 is calculated as 15.00 degrees - 10.00 = 5.00 degrees. Next, a normalization process is performed, dividing 5.00 / 40.00 by 0.125 to obtain the joint motion range parameter for axis 1 as 0.125. Similarly, for axis 2, with a starting position of -20.00 degrees, an ending position of -60.00 degrees, and a current position of -25.00 degrees, the calculated total range length is 40.00 degrees (absolute value), and the current displacement is 5.00 degrees (absolute value), resulting in a normalized parameter of 0.125. For axis 3, with a starting position of 0.00 degrees, an ending position of 30.00 degrees, and a current position of 6.00 degrees, the calculated normalized parameter is 0.200. For axis 5, with a starting position of 45.00 degrees and an ending position of 45.00 degrees, since the calculated total range length is 0.00, based on the above division-by-zero protection logic, its normalized parameter is directly set to 0.000. The calculations for all six axes are completed in this manner, ultimately generating a set of joint motion range parameters containing six normalized values.

[0026] Table 1 Calculation Table of Robot Joint Motion Range Parameters

[0027] As shown in Table 1, this step outputs a standardized set of joint motion range parameters through mathematical operations on physical angle data, providing a unified set of basic data for subsequent proportional mapping.

[0028] S102: Based on the joint motion interval parameter set, perform a proportional mapping operation according to the multi-axis normalized interval length and the corresponding stroke span, determine the state based on the mapping result and the interval start and end markers, and serialize and rearrange the occupancy markers according to the motion axis order to obtain the interval occupancy state sequence. After obtaining the joint motion range parameter set, the current motion progress of each axis is evaluated relative to its physical travel limit or planned travel to determine the motion state of each axis. First, a range start and end marker is set, which includes "Not Started" (corresponding to value 0), "In Progress" (corresponding to value range 0 to 1), and "Completed" (corresponding to value 1). The calculated normalized parameters of each axis are compared with the range start and end markers. Specifically, if the normalized parameter of an axis is equal to 0, its state is determined to be "Not Started"; if the normalized parameter is greater than 0 and less than 1, its state is determined to be "In Progress," and this normalized parameter is directly reserved as the occupancy ratio; if the normalized parameter is equal to or greater than 1, its state is determined to be "Completed." Based on the mapping results, further state determination and serialization rearrangement are performed. The determined state data is sorted according to the physical sequence number of each motion axis of the robot (i.e., axis 1 to axis 6). For axes in the "in progress" state, their normalized parameters are extracted as interval occupancy values; for axes that are "not started" or stationary (such as axis 5 in Table 1, where the start and end positions are the same), their interval occupancy values ​​are forcibly set to 0; for axes that are "completed," they are set to 1. During this process, a synchronization deviation threshold is introduced to determine the consistency of multi-axis coordinated motion. The threshold is set as follows: Through field testing, the time stamp difference between the arrival times of each axis at the target point is collected in 100 standard labeling actions, and its standard deviation is calculated to be 5 milliseconds. Based on the principle of 3 times the standard deviation, the synchronization deviation threshold is set to 15 milliseconds. Combined with the labeling process requirement of 120 pieces per minute, the total cycle of a single action is approximately 500 milliseconds. Therefore, the joint motion ratio corresponding to a time deviation of 15 milliseconds is approximately 15 / 500 = 0.03 (i.e., a 3% travel deviation). Continuing with the example data from S101, the data in Table 1 is processed: the parameter 0.125 for axis 1 is between 0 and 1, indicating "in progress," with an occupation value of 0.125; the parameter 0.200 for axis 3 is also indicated as "in progress," with an occupation value of 0.200; the parameter for axis 5 is 0.000, indicating "stationary / not started," with an occupation value of 0.000. At this point, the occupation ratio of axis 3 (0.200) is detected to be significantly higher than that of axis 1 (0.125), and the difference of 0.075 exceeds the proportion tolerance of 0.005 calculated based on the synchronization deviation threshold. This indicates that the movement progress of axis 3 is ahead of schedule and requires further adjustment. Finally, following the order of axes 1 to 6, the occupation values ​​of each axis (0.125, 0.125, 0.200, 0.100, 0.000, 0.100) are arranged sequentially to construct a sequence of interval occupation states. This sequence accurately reflects the actual completion distribution of each joint on the planned path at the current moment, revealing the asynchronous motion between axis 3 and other axes.

[0029] S103: Call the interval occupancy state sequence, perform linear interpolation operation based on the multi-axis occupancy interval ratio and the corresponding joint current position value, align and reorganize the interpolation results according to time, and associate the path order of the joint position arrangement to generate the initial path point set; The generated interval occupancy state sequence is invoked, and linear interpolation is performed in conjunction with the current position values ​​of each joint to generate smooth path points. This process first determines the target step size for interpolation. Based on the labeling process requirement of 120 pieces per minute, the time resolution for path planning is set to 10 milliseconds. The occupancy ratio of each axis at the current moment is read, and the remaining motion interval is calculated based on the dominant axis (i.e., the axis with the largest occupancy ratio, which in this example is axis 3, ratio 0.200). The remaining interval of axis 3 (1-0.200=0.800) is divided into several interpolation nodes. For example, if the plan is to complete the remaining motion within the next 100 milliseconds, it is divided into 10 time steps. For each time step, linear interpolation is performed. To eliminate accumulated errors and unify the calculation standard, the calculation logic adopts an absolute ratio mapping method: target position = motion start position + (total joint motion interval length × target occupancy ratio). A specific example is: for the lagging axis 1 (current ratio 0.125), the goal is to synchronize it with axis 3 (current ratio 0.200) at the endpoint. Calculate the theoretical position of axis 1 at the next time step (set the time step increment to 0.01). Assuming the time step corresponding to the planned next path point is 0.130 of the total journey (i.e., moving forward 0.005 or as set according to the step size), the interpolation is: 10.00 degrees (starting position) + 40.00 (total interval length) × 0.130 (target scale) = 15.20 degrees. Similarly, for axis 3, if the synchronous planning is to 0.205, the calculation is: 0.00 (starting position) + 30.00 (total interval length) × 0.205 (target scale) = 6.15 degrees. Combine the calculated angle values ​​of the six axes at the same time step into a coordinate vector, i.e., (15.20, -25.20, 6.15, 99.45, 45.00, 1.05). Subsequently, perform time alignment reorganization and joint position arrangement. The calculated single-point coordinate vectors are stored sequentially into the path array according to the time step sequence (t1, t2, t3...). During the storage process, path sequence association is performed to ensure that the joint positions of each axis of the subsequent path point satisfy the displacement continuity constraint relative to the previous path point. For example, it is checked whether the difference of 0.20 degrees between the position of axis 1 at time t1 (15.20 degrees) and time t0 (15.00 degrees) is within the allowable range of the motor's maximum angular velocity (e.g., no more than 0.5 degrees per 10 milliseconds). If the check passes, the coordinate vector is confirmed as a valid path point. Finally, the output contains an initial path point set containing a series of six-dimensional coordinate vectors, which covers the complete trajectory from the current position to the final position of the motion.

[0030] Please see Figure 3 The specific steps of S2 are as follows: S201: Obtain the joint position combination corresponding to the multi-path points in the initial path point set, perform combination mapping on the multi-motion axis joint position values ​​for the multi-path point index, and vectorize and rearrange the multi-motion axis joint positions under the same temporal index to generate a path point joint combination sequence. Based on the generated initial path point set, which stores discretized full data from the start point to the end point of motion using a time series index, a data extraction operation is performed for each discrete time node, i.e., the path point index. The joint position values ​​of the multiple motion axes under the current index are identified; for example, at the time index of the first node, provisional angle values ​​for six joint axes are read. To achieve data standardization, a vectorized rearrangement instruction is invoked, loading the discretely read scalar angle data into a six-dimensional column vector according to the physical connection order of the ABB IRB1600 robot's joints, i.e., from base axis 1 to end flange axis 6. During this process, data validity is verified for each joint position value to ensure that the value does not exceed the robot's soft limit range. With a practical example, when processing path point index P1 at time 10 milliseconds, the joint position data read is as follows: Axis 1 is 15.20 degrees, Axis 2 is -25.20 degrees, Axis 3 is 6.15 degrees, Axis 4 is 99.45 degrees, Axis 5 is 45.00 degrees, and Axis 6 is 1.05 degrees. These six values ​​are sequentially filled into the vector space. The sequence contains data items of 15.20, -25.20, 6.15, 99.45, 45.00, and 1.05, respectively. For Axis 5, since its starting and ending positions are both 45.00 degrees, it is in a static state, and its value remains constant in the sequence. Next, index P2 is processed, corresponding to time 20 milliseconds. After linear interpolation, the read data is set to Axis 1 is 15.40 degrees, Axis 3 is 6.30 degrees, etc., and the rearrangement is performed again to generate the corresponding vector sequence. The above combination mapping and vectorized rearrangement operations are repeated for approximately 100 discrete points in the entire path point set. During the rearrangement process, an index mapping table is simultaneously established, strictly binding each generated joint combination sequence with a timestamp to form a path point joint combination sequence.

[0031] S202: Based on the path point joint combination sequence and the interval occupancy status sequence, compare the interval occupancy ratios corresponding to multiple motion axes within multiple path points, judge the difference between the multi-axis occupancy ratios under the same path point and the preset ratio benchmark value, and sort them to obtain a motion axis occupancy ratio comparison identifier set. The generated pathpoint joint combination sequence is retrieved and combined with the determined interval occupancy status sequence to perform real-time occupancy ratio calculation and comparison for multi-axis intervals. For each pathpoint, the instantaneous normalized occupancy ratio of each axis at the current moment is calculated backward using the joint angle values ​​in the current combination sequence. The calculation logic is as follows: First, the total joint motion interval length corresponding to each axis is checked. If the length value is 0 (i.e., the stationary axis whose starting position equals the ending position), its instantaneous occupancy ratio is directly set to 0 to avoid the calculation error of dividing by zero. If the length value is not 0, the current joint angle value is subtracted from the motion starting position, and the difference is divided by the total joint motion interval length. Subsequently, a "proportional reference value" is introduced as a reference line for measuring multi-axis synchronization. The proportional reference value is set based on the arithmetic mean of the occupancy ratios of all non-stationary motion axes in the "in progress" state at the current moment. The instantaneous ratios of all active axes are traversed, summed, and divided by the number of active axes to obtain the proportional reference value at that moment. Based on this reference value, the difference judgment operation is performed. The instantaneous occupancy ratio of each motion axis is subtracted from the proportional baseline value to obtain the proportional deviation value. The motion axes are then sorted according to the algebraic magnitude of the deviation values. A specific example is provided below using the implementation data shown in Table 2. At path point P1, the instantaneous occupancy ratio of each axis is calculated: Axis 1 is calculated as (15.20−10.00) / 40.00=0.130; Axis 2 is calculated as ((−25.20)−(−20.00)) / (−40.00), and the absolute value is removed by dividing by the total interval, resulting in 0.130; Axis 3 is calculated as (6.15−0.00) / 30.00=0.205; Axis 4 and Axis 6 are calculated similarly and both are 0.105. For Axis 5, since its total interval length is 0, its ratio is directly determined to be 0 according to the above protection logic, classifying it as a stationary axis. Active axes are identified as axes 1, 2, 3, 4, and 6, a total of 5. Calculate the scale baseline value: Add the scales of the five axes (0.130, 0.130, 0.205, 0.105, 0.105) to get 0.675, then divide by 5 to get 0.135. Next, calculate the deviation: Axis 1 deviation is 0.130 - 0.135 = -0.005; Axis 3 deviation is 0.205 - 0.135 = 0.070; Axis 4 deviation is 0.105 - 0.135 = -0.030. Sort the deviation values ​​from largest to smallest: Axis 3 (0.070), Axis 1 (-0.005), Axis 2 (-0.005), Axis 4 (-0.030), Axis 6 (-0.030). This result shows that the movement progress of Axis 3 is significantly ahead of the average level, while Axis 4 and Axis 6 are relatively lagging behind.

[0032] Table 2 Comparison and Ranking of Multi-Axis Occupancy Ratio

[0033] Referring to Table 2, a quantitative assessment of the synchronization of multi-axis motion was completed, and axis 3 was identified as the motion axis with the fastest progress at the current moment. This data directly determines the subsequent selection of the dominant joint.

[0034] S203: Based on the motion axis occupancy ratio comparison identifier set, perform ratio filtering and judgment for multi-path points, select the motion axis with the optimal occupancy ratio sorting position, and perform constraint mapping by combining the selected motion axis index with the corresponding joint position to generate dominant joint constraint data; Based on the generated set of motion axis occupancy ratio comparison identifiers, the "dominant joint" that plays a decisive role in the overall motion rhythm is determined. A filtering logic is executed: the motion axis with the highest ranking, i.e., the largest and positive ratio deviation, is selected as the dominant motion axis. The advantage of this approach is that by locking the joint with the fastest progress or the highest interval occupancy ratio as a benchmark, the entire robot's motion time axis is aligned to this "leading" joint, thus ensuring, at the physical level, that all joints can reach the target position at the same time, avoiding micro-pauses or jitters caused by fast axes waiting for slow axes. After determining the dominant motion axis, a constraint mapping operation is performed. The index of the dominant motion axis and its corresponding joint position combination are extracted to generate dominant joint constraint data. This data includes not only the current angle of the dominant axis but also its corresponding normalized ratio, which is established as the "global reference progress" for the current time step. This global reference progress is forcibly assigned to other non-dominant axes as the theoretical progress target they should follow in their optimization paths. Following the previous example, at path point P1, axis 3 is detected to have a sorting position of 1 and a deviation value of 0.070, therefore axis 3 is selected as the dominant motion axis. The current position of axis 3, 6.15 degrees, and its instantaneous occupancy ratio of 0.205 are extracted. The generated dominant joint constraint data is defined as follows: the globally normalized progress at time node t=1 of the optimized path should be corrected to 0.205. This means that although axis 1 was originally planned to be at a progress of 0.130 (i.e., 15.20 degrees) at this moment, due to the constraint of dominant joint axis 3, axis 1 must replan its trajectory to reach or match the progress requirement of 0.205 at the corresponding time node. Specifically, the 0.205 ratio is mapped back to the motion range of axis 1: first, the product of the total motion range length of axis 1, 40.00 degrees, and 0.205 is calculated, resulting in 8.20 degrees; then, this result is added to the starting position of axis 1, 10.00 degrees, to obtain 18.20 degrees. The calculated 18.20 degrees is the corrected target position of axis 1 under the dominant constraint. The above filtering and mapping process is repeated for all path points. For each time step, a dominant axis is identified, and a constraint data package containing the target angles of the six axes is generated based on the proportion of that dominant axis. Finally, the complete dominant joint constraint dataset is output.

[0035] Please see Figure 4 The specific steps of S3 are as follows: S301: Obtain the proportion of the remaining joint motion range for the corresponding motion axis based on the dominant joint constraint data, call the joint motion range parameter for the dominant motion axis, perform proportion calculation on the range length value, and arrange the calculation results according to the path point index to generate a sequence of remaining joint range proportions. The generated dominant joint constraint data is retrieved, and the index identifier marked as the dominant motion axis is extracted from it. Based on the previous example, this is axis 3, and the normalized occupancy ratio of this axis at the current path point is 0.205. The processor performs a quantitative calculation of the remaining process based on this dominant ratio. By subtracting the current normalized occupancy ratio of 0.205 from the unit value of 1.000 defined for the entire stroke, the remaining joint motion range ratio of the dominant axis is obtained as 0.795. This value represents the relative share that all cooperating motion axes should have remaining in physical stroke to complete the labeling task under ideal synchronization. To construct a complete path planning constraint channel, the ratio is not only calculated at a single time point, but also batched for a preset look-ahead window, such as the indices of the next 10 path points. The planned positions of the dominant axes at subsequent path points are retrieved sequentially. At the next path point index P2, the planned position of dominant axis 3 is set to 6.30 degrees. The generated joint motion range parameter set is retrieved to obtain the total range length of axis 3 (30.00 degrees) and the starting position (0.00 degrees). The occupancy ratio at P2 is calculated to be 0.210. Subsequently, the remaining proportion at P2 is calculated to be 0.790. This calculation logic is repeated until the entire lookahead window is covered. After the proportion calculation is complete, a serialization and arrangement operation is performed. The calculated series of remaining proportion values ​​are stored in the cache according to the time order of the path point indices, generating a sequence of remaining joint interval proportions. This sequence, by quantifying the "margin" of the dominant axis, establishes a unified reduction rhythm that the entire multi-axis must follow in subsequent movements.

[0036] S302: Based on the remaining joint interval ratio sequence, obtain the joint motion interval parameters corresponding to the other motion axes in the path planning, use the remaining joint interval ratio sequence to perform boundary conversion on the total joint motion interval length of the multi-motion axis, calculate the effective interval boundary of the other motion axes under the current ratio, and obtain the reverse clipping interval boundary set; Based on the generated sequence of remaining joint interval proportions, a reverse mapping is performed on the non-dominant axes, namely axes 1, 2, 4, and 6, to determine their physical motion boundaries. First, the joint motion interval parameters of the remaining axes are retrieved from memory, including the start position, end position, and total interval length of each axis. For the remaining proportion value under each pathpoint index, a "synchronization tolerance coefficient" is introduced to define a reasonable motion bandwidth. The process of setting the synchronization tolerance coefficient is experimentally calibrated based on the servo response characteristics of the ABBIRB1600 robot. In the experiment, different coefficient values ​​were set, ranging from 0.005 to 0.050, and 100 full-speed labeling tests were performed, monitoring the consistency between the end effector jitter amplitude and arrival time. Experimental data shows that when the coefficient is set to 0.020, i.e., 2% of the total travel deviation, the end effector positioning accuracy is stable within ±0.05 mm without significant overshoot. Therefore, the synchronization tolerance coefficient is fixed at 0.020. Based on this coefficient, a reverse mapping calculation is performed for axis 1. Taking the current dominant axis occupancy ratio of 0.205, the target synchronization ratio range for axis 1 is determined to be between 0.205 minus 0.020 and 0.205 plus 0.020, i.e., 0.185 to 0.225. Then, the total interval length of axis 1 (40.00 degrees) and the starting position (10.00 degrees) are used to convert the proportional boundaries into physical angle boundaries. The lower limit is calculated as 10.00 + 40.00 * 0.185 = 17.40 degrees; the upper limit is calculated as 10.00 + 40.00 * 0.225 = 19.00 degrees. The above calculation is repeated for all non-dominant axes, with a consistency check performed. The check logic is: check if the calculated physical boundary is within the robot's soft limit range, for example, ±180 degrees for axis 1. If the calculated boundary exceeds the limit, the boundary value is truncated to the limit value. Finally, the lower and upper limits of the physical angles at each path point for each axis are integrated to obtain the reverse trimming interval boundary set. The results indicate that, in order to maintain synchronization with axis 3, the trajectory of axis 1 must be limited to between 17.40 and 19.00 degrees. If the current actual position of axis 1 is 15.20 degrees, it indicates that axis 1 is severely lagging and must move towards this range with maximum acceleration in subsequent planning.

[0037] S303: Based on the reverse clipping interval boundary set, aggregate multi-motion axis clipping intervals for multi-path point indexes, perform joint constraint arrangement on the clipped intervals and the intervals corresponding to the dominant motion axis, and perform interval continuity verification on the arrangement results to generate path joint constraint range; Based on the generated set of reverse clipping interval boundaries, multidimensional data aggregation is performed on the multipath point index to construct a "feasible pipeline" in a six-dimensional joint space. For each time step, the planned position of the dominant motion axis itself is treated as a very small interval, such as 6.15 degrees ± 0.01 degrees, and combined with the clipping intervals of other non-dominant axes, such as 17.40 to 19.00 degrees for axis 1, forming a six-dimensional hypercube constraint box. On this basis, a rigorous interval continuity check is performed. This check aims to ensure that the constraint range between adjacent path points meets the robot's dynamic limits. The rated maximum angular velocity parameters of each joint of the robot are read, for example, 150 degrees / second for axis 1, and the time step of the path point is 10 milliseconds. The difference between the center points of the constraint intervals of two adjacent time steps is calculated. The check logic is: if the result of the position difference divided by the time step is greater than the maximum angular velocity, the constraint interval is determined to violate dynamic continuity. At this time, dynamic correction is performed to move the constraint interval of the next time step closer to the interval of the previous time step, reducing the unreachable area. Taking axis 1 as an example, the current position is 15.20 degrees, while the target lower limit is 17.40 degrees. The position difference is 2.20 degrees. If the time step is 10 milliseconds, the required velocity is 220 degrees / second, which exceeds the limit of 150 degrees / second. Therefore, the maximum feasible displacement calculated based on the maximum velocity is 150 * 0.01 = 1.50 degrees. The actual effective constraint range of axis 1 in the next moment is corrected to an upper limit of 15.20 + 1.50 = 16.70 degrees. Combined with the synchronization direction, the final corrected constraint range is 16.50 to 16.70 degrees, taking the side closer to the synchronization target.

[0038] Please see Figure 5 The specific steps of S4 are as follows: S401: Based on the joint position combination corresponding to the multi-path points in the initial path point set, collect the multi-joint position values ​​under the multi-path point index, align the joint position values ​​with the corresponding joint dimension index, and perform structured aggregation to generate a path point joint position combination set. First, a full scan of the initial pathpoint set in memory is initiated. This initial pathpoint set contains the generated raw motion trajectory data. Discrete pathpoint data packets are extracted one by one based on the timestamp index. For each pathpoint index, encoder values ​​or planned values ​​for the six joint axes are called in parallel to perform data cleaning and alignment operations. A six-dimensional array structure is created as a container for a single pathpoint. Following the joint physical sequence numbers of the ABB IRB 1600 robot, i.e., axes 1 to 6, the read floating-point angle values ​​are sequentially filled into the corresponding dimensions. During this process, structured aggregation instructions are executed to integrate the discrete scalar data into vectorized objects. Simultaneously, metadata tags, including timestamps, velocity feedforward values, and acceleration feedforward values, are added to each aggregated vector to ensure the integrity of the data's physical attributes. Taking path point P1, corresponding to a time of 10 milliseconds, as an example, the initial planned positions of each axis are read from memory: axis 1 is 15.20 degrees, axis 2 is -25.20 degrees, axis 3 is 6.15 degrees, axis 4 is 99.45 degrees, axis 5 is 45.00 degrees, and axis 6 is 1.05 degrees. The format of these values ​​is verified, for example, to ensure they are double-precision floating-point numbers, and they are written into a vector in the order of 15.20, -25.20, 6.15, 99.45, 45.00, and 1.05. For path point P2, corresponding to a time of 20 milliseconds, the values ​​are read: axis 1 is 15.40 degrees, axis 3 is 6.30 degrees, etc., and the corresponding vectors are generated. This alignment and aggregation operation is repeated for approximately 100 sampling points along the entire path, generating an ordered list containing 100 six-dimensional vectors, i.e., the set of path point joint positions.

[0039] S402: Based on the set of path point joint positions and the range of path joint constraints, perform interval assignment judgment on each dimension of the joint position vector, mark the path point index that exceeds the interval boundary as invalid, and obtain the set of path points within the constraints; The generated set of pathpoint joint positions is invoked, and the generated path joint constraint range is loaded. For each pathpoint index, the interval attribution judgment logic is initiated. The joint angle value of each dimension in the set of joint positions is extracted one by one and compared with the physical lower limit and physical upper limit within the constraint range after correction for maximum angular velocity. The specific comparison action is as follows: it is determined whether the currently extracted joint angle value is greater than or equal to the physical lower limit and less than or equal to the physical upper limit. If both conditions are met, the value of that dimension is determined to be within the valid range and marked as "compliant"; if either condition is not met, such as the value being less than the lower limit or greater than the upper limit, the value is determined to be outside the interval boundary, and the joint dimension under the current pathpoint index is marked as "invalid". Continuing with the previous example, at pathpoint P1, the initial position value of axis 1 is extracted as 15.20 degrees. The calculated and corrected maximum angular velocity constraint range for axis 1 is 16.50 to 16.70 degrees (Note: the theoretical synchronization range was originally 17.40 to 19.00 degrees, but was dynamically pruned by S303 due to the 150 degrees / second speed limit). Judgment: 15.20 degrees is less than the corrected lower limit of 16.50 degrees, therefore axis 1 is determined to be in an invalid state at time P1, indicating that even after considering the dynamic boundary, this position still lags behind the effective synchronization rhythm of the system. Similarly, axis 3 (the dominant axis) is checked; its position of 6.15 degrees is within the constraint range of 6.14 to 6.16 degrees, and is deemed compliant. This traversal check is performed on all joint axes of all path points. For any path point where any joint axis is marked as "invalid," the overall state of that path point is set to "to be optimized," and the specific deviation is recorded. After a full path scan, a set of path points within the constraints is generated. This set is actually a dataset with a state mask, clearly distinguishing which joints at which time points deviated from the dynamically corrected synchronization trajectory. For example, in tabular form, at time P1 (10ms), the initial value of axis 1 is 15.20 degrees, with a constraint range of 16.50 to 16.70 degrees, which is considered excessive (too low) and marked as invalid; while the initial value of axis 3 is 6.15 degrees, within the constraint range of 6.14 to 6.16 degrees, and is considered compliant. At time P2 (20ms), the initial value of axis 1 is 15.40 degrees, with the constraint range also using the corresponding dynamically corrected value (e.g., 18.00 to 18.20 degrees), again considered excessive (too low) and marked as invalid. At time P50 (500ms), the initial value of axis 1 is 25.00 degrees, with a constraint range of 24.80 to 25.20 degrees, and is considered compliant.

[0040] S403: Based on the set of path points within constraints, perform a filtering and reorganization operation on the path points corresponding to the initial set of path points, arrange the filtered path points according to the original path order, and generate an optimized set of path points; Based on the generated set of constrained path points, path point filtering and reorganization operations are performed to construct an optimization target set for the final path planning. The path point set with status markers is traversed, and data replacement logic is executed for path point indices marked as "invalid." The original joint position values ​​that were determined to be non-compliant (e.g., 15.20 degrees for axis 1) are no longer retained. Instead, the "reverse clipping interval" generated in S303 is directly mapped to the index position, replacing the determined scalar coordinates with the search space of feasible solutions, such as 17.40 to 19.00 degrees. For path points marked as "compliant," such as at time P50, their original joint position values ​​are retained as fixed anchor points in path planning. The processed data is rearranged according to the original time series index, i.e., from P1 to Pend. For invalid points, the new data structure contains the allowed joint angle range at that time; for compliant points, it contains the determined angle values. This process essentially reconstructs a single linear trajectory into a "time-varying pipeline." In sections with poor synchronicity (such as acceleration sections), the pipeline widens to allow the RRT* algorithm to search for the optimal solution within a range; in sections with good synchronicity, the pipeline shrinks to specific points to maintain trajectory stability. In the example, for index P1, the generated optimized data items are: axis 1 ranging from 17.40 to 19.00 degrees, axis 3 value of 6.15 degrees, and the remaining axes corresponding to their respective ranges or values. For index P2, axis 1 ranging from 17.65 to 19.25 degrees is generated. These data items, which mix scalar and interval values, are then encapsulated to generate the optimized path point set.

[0041] Please see Figure 6 The specific steps of S5 are as follows: S501: Based on the optimized path point set, obtain the multi-joint position values ​​corresponding to multiple path points, and perform mapping between joint numbers and position values ​​to form a joint position vector sequence. Then, perform structured combination operations to generate a path point joint position sequence. Based on the generated optimized pathpoint set, a deep analysis program is first initiated on the hybrid data structure contained in this set in memory. This set includes fixed joint value points marked as compliant and undetermined range points marked as invalid and reorganized into intervals. A full extraction operation is performed on the data packets under each pathpoint index. For pathpoints set as fixed values, their scalar angle values ​​are directly read. For pathpoints set as range intervals, the RRT* algorithm is called to randomly sample within the feasible region channel defined by S403 and which must strictly conform to the dynamic correction of S303. The optimal cost point is selected as the deterministic coordinate at that moment using the cost function. For example, for pathpoint P1 at moment 1, although its theoretical synchronization requirement is 17.40 to 19.00 degrees, due to the maximum angular velocity constraint, the actual effective constraint range has been corrected by S303 to 16.50 to 16.70 degrees. Therefore, the RRT* algorithm samples within this correction interval and, after evaluation by the cost function, selects 16.65 degrees as the specific execution position at that moment (this position has a displacement of 1.45 degrees relative to the previous moment's 15.20 degrees, corresponding to a speed of 145 degrees / second, which is less than the limit of 150 degrees / second, thus conforming to physical constraints). A mapping table between joint physical dimensions and logical control channels is then established. The extracted or selected six joint angle values ​​are strictly aligned according to the hardware connection order of axes 1 to 6 of the ABBIRB1600 robot, constructing a six-dimensional column vector. Simultaneously, the kinematic parameter library is called to attach corresponding metadata tags for maximum speed limits, maximum acceleration limits, and torque estimates to each dimension of this vector, ensuring that each data point contains not only positional information but also dynamic constraint attributes. The numerical extraction, alignment mapping, and attribute appending process is repeated for approximately 100 to 200 discrete time steps along the entire path. The resulting 100 six-dimensional vectors are then pushed into a dynamic array or deque in the order of timestamps t0 to tend. During this process, structured combination operations are performed to check the consistency of vector dimensions and the double-precision floating-point format of the data type. This transforms the discrete original coordinate points, which may contain interval definitions, into a path point joint position sequence with complete kinematic properties and a uniform format.

[0042] S502: Based on the path point joint position sequence, perform a sequence consistency check on adjacent path points, determine the sequence arrangement state according to the path point index increment relationship, and remove index positions that do not conform to the increment relationship to obtain a sequence of joint positions with consistent order. Based on the generated pathpoint joint position sequence, a sequence integrity and timing logic verification engine is instantiated. This engine aims to eliminate non-monotonic time nodes that may be caused by randomness in the sampling algorithm, multi-threaded write conflicts, or interpolation calculation errors. A two-pointer traversal is initialized, pointing to the current verification pathpoint. Compared with the previous valid path point Read the timestamp index values ​​corresponding to the two points and perform a strict incrementing relationship check, with the condition being: Is it strictly greater than And time difference If a path point is found during the traversal, and the preset sampling period is 10 milliseconds. The index is the same as or less than the previous point, for example... The corresponding time is 50ms and then... The corresponding time is 48ms. For points with temporal reversals or redundant overlaps, an immediate removal mark operation is performed on the index position to remove it from the main motion sequence to prevent the servo controller from receiving backtracking instructions. At the same time, static detection is performed in the spatial dimension to calculate the Euclidean distance or joint difference norm between adjacent points. If the index increases but all six-axis positions do not change significantly (the change is less than the encoder resolution), it is determined to be an invalid stop point and is removed to optimize the motion beat. After completing a single scan and cleaning of the entire sequence, a memory defragmentation operation is performed to relink all valid nodes that have not been marked and removed according to the original topological order, filling the index gaps left by the removed points. Through this strict monotonic filtering and redundancy cleaning, a sequentially consistent joint position sequence that is strictly increasing in the time dimension and has no redundant pauses in the spatial dimension is output.

[0043] S503: Based on the sequential joint position sequence, calculate the joint displacement continuity metric for adjacent joint position vectors, retain the path point sequence located at the joint displacement continuity threshold and perform path connection operation to generate an optimized robot motion path; Based on the generated sequential joint position sequence, the final trajectory smoothness evaluation and connection stage begins. For each pair of adjacent joint position vectors in the sequence, joint angle values ​​are extracted dimension by dimension. For example, the joint angle values ​​for axis 1 are extracted. Time and The system calculates the angle value at each moment and calls a preset joint displacement continuity measurement model to perform micro-smoothness calculation. This calculation process aims to quantify the degree of dynamic transition between adjacent points and combine it with the inherent clearance characteristics of the machine for comprehensive scoring. The calculated measurement value is compared with a preset continuity threshold. If the measurement value is less than or equal to the threshold, the path segment is determined to meet the smooth transition requirements. The path point is retained and connected to the previous point by fitting a fifth-order polynomial or B-spline curve. If the measurement value exceeds the threshold, a local interpolation correction mechanism is triggered to insert a transition point between the two points or adjust the time step to reduce the speed requirement. Finally, all the verified discrete points are fitted into a continuous and differentiable spline curve to generate an optimized robot motion path.

[0044] The formula for calculating the continuity metric of joint displacement is: ; Detailed explanation of formula parameters and operational logic: In the formula... It represents a dimensionless scalar that measures the continuity of joint displacement and is used to quantitatively evaluate whether the connection between path points meets the smoothness requirements. Represents the corresponding joint dimension at the current moment. The joint position vector values ​​(unit: degrees); This represents the corresponding joint dimension at the previous moment. The joint position vector values ​​(in degrees), and the difference between the two. The absolute value operation was used to obtain the pure displacement magnitude, reflecting the requirements of dynamic programming; This represents a displacement compensation reference value that is independent of adjacent differential calculations. This value corresponds to the inherent backlash or repeatability error of the robot's deceleration mechanism. This represents the weighted adjustment coefficient, used to adjust the weight of the static error term in the overall evaluation; This term constructs a nonlinear static damping term through the calculation of squaring and weighted square root. Its purpose is to introduce a "minimum motion cost". Even the smallest displacement must overcome this basic threshold to prevent micro-oscillations caused by numerical noise. This represents the normalization factor, typically taken as the maximum allowable physical displacement of the joint within that time step, calculated using division. The physical displacement is converted into a proportional value relative to the ultimate capability, facilitating standardized comparison with a uniform threshold (e.g., 1.0). Parameter acquisition and assignment process: and The sequence is obtained by reading the "sequential joint position sequence" output by S502 in real time. In this embodiment, the time index of axis 1 is extracted. (100ms) and Data from 90ms. Monitoring showed the position at the previous moment. Degree, current position Spend. (Displacement Compensation Reference Value): This parameter was obtained through an ISO9283 repeatability test on the ABBIRB1600 robot. In the experiment, the robot was controlled to repeatedly reach the same target point 100 times, and the standard deviation of the actual encoder feedback value was measured. Experimental data shows that the position repeatability is approximately... mm, which translates to approximately the angle of joint 1 on axis 1. Degree. Therefore, set . (Weighted Adjustment Coefficient): This coefficient is set based on the servo's rigidity. By applying a step signal to axis 1 and analyzing the frequency response, a gain coefficient of [value missing] is selected to suppress low-frequency creep ripple. .Right now . (Normalization factor): This parameter is based on the maximum angular velocity of the joint. and time step ( ) Calculation. The theoretical maximum displacement is To provide a certain dynamic margin as a safety boundary, the normalized reference value is set to [value missing]. Degree. That is. Substitute into the formula : Calculate the absolute displacement difference of the joint positions: Calculate the internal product of the static compensation terms: First, calculate the square of the baseline value: Then multiply by the weighting factor: Take the square root of the static compensation term: Calculate the sum of the numerators: Perform normalized division: This result indicates that in At time 1, the calculated continuity metric value Less than the preset baseline threshold This indicates that the displacement change is still within a safe range after considering mechanical backlash and dynamic limitations, and the execution path will be connected; while Time, metric If the threshold is exceeded, it means that the instruction may cause the motor to overload or the movement to be unstable, and the point will be determined to be subject to interpolation smoothing.

[0045] Please see Figure 7 A multi-axis robot path planning and optimization system, comprising: The path generation module collects the current joint position, start and end position of motion of the joints corresponding to the multi-axis motion axis of the double-sided synchronous labeling multi-axis robot, calculates the joint motion interval distance and the total motion interval length and converts them into the interval occupancy ratio, performs linear interpolation with the start and end positions of motion, generates an initial path point set and passes it to the main analysis module. The dominant analysis module obtains the joint position group corresponding to the multipath points in the initial path point set, performs a comparison of the interval occupancy ratio of the motion axes within the multipath points based on the interval occupancy status, selects the motion axis with the optimal ratio, generates dominant joint constraint data, and transmits it to the interval constraint module. The interval constraint module obtains the proportion of the remaining joint motion interval of the corresponding motion axis based on the dominant joint constraint data, extracts the total joint motion interval length, performs boundary conversion according to the proportion of the remaining joint motion interval, generates the path joint constraint range, and passes it to the path filtering module. The path filtering module determines whether the path points are within the path joint constraint range based on the joint position combination corresponding to the multi-path points in the initial path point set, filters the path points, generates an optimized path point set, and passes it to the motion optimization module. The motion optimization module, based on the optimized path point set, obtains the joint positions corresponding to multiple path points and performs path point sequence consistency verification and joint displacement continuity judgment to generate an optimized robot motion path.

[0046] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A path planning and optimization method for a multi-axis robot, characterized in that, Includes the following steps: S1: Collect the current position of the joints corresponding to the multiple motion axes of the double-sided synchronous labeling multi-axis robot, the start and end positions of the motion, calculate the joint motion interval distance and the total motion interval length and convert them into the interval occupancy ratio, perform linear interpolation with the start and end positions of the motion to generate the initial path point set; S2: Obtain the joint position group corresponding to the multi-path points in the initial path point set, compare the interval occupancy ratio of the motion axes in the multi-path points with the interval occupancy status, select the motion axis with the optimal ratio, and generate dominant joint constraint data; S3: Based on the dominant joint constraint data, obtain the proportion of the remaining joint motion range of the corresponding motion axis, extract the total joint motion range length of the corresponding joint, perform boundary conversion according to the proportion of the remaining joint motion range, and generate the path joint constraint range. S4: Based on the joint position combination corresponding to the multi-path points in the initial path point set, determine whether it is within the path joint constraint range and filter the path points to generate an optimized path point set; S5: Based on the optimized path point set, obtain the joint positions corresponding to the multiple path points and perform path point sequence consistency verification and joint displacement continuity judgment to generate an optimized robot motion path.

2. The multi-axis robot path planning optimization method according to claim 1, characterized in that, The initial path point set includes a path point index sequence, multi-axis interpolated joint positions, and interval occupancy status. The dominant joint constraint data includes the dominant motion axis identifier, the corresponding joint interval occupancy ratio, and constraint priority parameters. The path joint constraint range includes the upper and lower bounds of motion axis joint clipping, the remaining motion interval ratio mapping, and the set of feasible joint intervals. The optimized path point set includes path points within constraints and combinations of joint positions. The optimized robot motion path includes the joint position temporal trajectory, the path point sequence, and the continuous change sequence of joint displacements.

3. The method for optimizing path planning of a multi-axis robot according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: Collect the current position, start and end position of the joints corresponding to the multi-axis motion of the double-sided synchronous labeling multi-axis robot, calculate the total joint motion range length, perform difference calculation on the current position and start and end position of the multi-axis joints, perform normalization processing on the multi-axis span based on the maximum stroke range of the multi-axis, and generate a set of joint motion range parameters. S102: Based on the joint motion interval parameter set, perform a proportional mapping operation according to the multi-axis normalized interval length and the corresponding stroke span, determine the state based on the mapping result and the interval start and end markers, and serialize and rearrange the occupancy markers according to the motion axis order to obtain the interval occupancy state sequence. S103: Call the interval occupancy state sequence, perform linear interpolation operation according to the multi-axis occupancy interval ratio and the corresponding joint current position value, align and reorganize the interpolation results according to time, and associate the path order of the joint position arrangement to generate an initial path point set.

4. The multi-axis robot path planning optimization method according to claim 1, characterized in that, The specific steps of S2 are as follows: S201: Obtain the joint position combination corresponding to the multi-path points in the initial path point set, perform combination mapping on the multi-motion axis joint position values ​​for the multi-path point index, and vectorize and rearrange the multi-motion axis joint positions under the same temporal index to generate a path point joint combination sequence. S202: Based on the path point joint combination sequence and the interval occupancy status sequence, perform a comparison of the interval occupancy ratio corresponding to multiple motion axes within multiple path points, and judge and sort the difference between the multi-axis occupancy ratio under the same path point and the preset ratio benchmark value to obtain a motion axis occupancy ratio comparison identifier set. S203: Based on the set of motion axis occupancy ratio comparison identifiers, perform ratio filtering and judgment for multi-path points, select the motion axis with the optimal occupancy ratio sorting position, and perform constraint mapping between the selected motion axis index and the corresponding joint position to generate dominant joint constraint data.

5. The multi-axis robot path planning optimization method according to claim 4, characterized in that, The proportional benchmark value is determined by summarizing the interval occupancy ratios of multiple motion axes in the interval occupancy state sequence within the initial path point set, collecting the occupancy ratio values ​​of the same type of motion axis under the multi-path point index, and performing an average calculation on the value set.

6. The method for optimizing path planning of a multi-axis robot according to claim 1, characterized in that, The specific steps for S3 are as follows: S301: Obtain the proportion of the remaining joint motion interval for the corresponding motion axis based on the dominant joint constraint data, call the joint motion interval parameter for the dominant motion axis, perform proportion calculation on the interval length value, and arrange the calculation results according to the path point index to generate a sequence of remaining joint interval proportions. S302: Based on the remaining joint interval ratio sequence, obtain the joint motion interval parameters corresponding to the other motion axes in the path planning, use the remaining joint interval ratio sequence to perform boundary conversion on the total joint motion interval length of the multi-motion axis, calculate the effective interval boundary of the other motion axes under the current ratio, and obtain the reverse clipping interval boundary set; S303: Based on the reverse clipping interval boundary set, aggregate multi-motion axis clipping intervals for multi-path point index, perform joint constraint arrangement on the clipped intervals and the intervals corresponding to the dominant motion axis, and perform interval continuity verification on the arrangement results to generate path joint constraint range.

7. The method for optimizing path planning of a multi-axis robot according to claim 1, characterized in that, The specific steps of S4 are as follows: S401: Based on the joint position combination corresponding to the multi-path points in the initial path point set, collect the multi-joint position values ​​under the multi-path point index, align the joint position values ​​with the corresponding joint dimension index and perform structured aggregation to generate a path point joint position combination set. S402: Based on the set of path point joint positions and the range of path joint constraints, perform interval assignment judgment on each dimension of the joint position vector, mark the path point index that exceeds the interval boundary as invalid, and obtain the set of path points within the constraints. S403: Based on the set of path points within the constraints, perform a filtering and reorganization operation on the path points corresponding to the initial set of path points, and arrange the filtered path points in the original path order to generate an optimized set of path points.

8. The method for optimizing path planning of a multi-axis robot according to claim 1, characterized in that, The specific steps of S5 are as follows: S501: Based on the optimized path point set, obtain the multi-joint position values ​​corresponding to the multi-path points, and perform mapping between the joint number and the position value to form a joint position vector sequence and perform structured combination operation to generate a path point joint position sequence. S502: Based on the path point joint position sequence, perform a sequence consistency check on adjacent path points, determine the sequence arrangement state according to the path point index increment relationship, and remove index positions that do not conform to the increment relationship to obtain a sequence of joint positions with consistent order. S503: Based on the sequential joint position sequence, calculate the joint displacement continuity metric for adjacent joint position vectors, retain the path point sequence located at the joint displacement continuity threshold and perform path connection operation to generate an optimized robot motion path.

9. The method for optimizing path planning of a multi-axis robot according to claim 8, characterized in that, The joint displacement continuity threshold is determined by collecting the displacement difference of the corresponding joint dimension under adjacent path points to form a set of joint displacement difference values, performing a sorting operation on the set of difference values, and extracting the median value of the multi-joint displacement difference distribution.

10. A multi-axis robot path planning and optimization system, characterized in that, The system is used to implement the multi-axis robot path planning optimization method according to any one of claims 1-9, and the system includes: The path generation module collects the current joint position, start and end position of motion of the joints corresponding to the multi-axis motion axis of the double-sided synchronous labeling multi-axis robot, calculates the joint motion interval distance and the total motion interval length and converts them into the interval occupancy ratio, performs linear interpolation with the start and end positions of motion, generates an initial path point set and passes it to the main analysis module. The dominant analysis module obtains the joint position group corresponding to the multi-path points in the initial path point set, performs a comparison of the interval occupancy ratio of the motion axes within the multi-path points in combination with the interval occupancy status, selects the motion axis with the optimal ratio, generates dominant joint constraint data, and transmits it to the interval constraint module. The interval constraint module obtains the proportion of the remaining joint motion interval of the corresponding motion axis based on the dominant joint constraint data, extracts the total joint motion interval length of the corresponding joint, performs boundary conversion according to the proportion of the remaining joint motion interval, generates the path joint constraint range, and passes it to the path filtering module. The path filtering module determines whether the path points are within the path joint constraint range based on the joint position combination corresponding to the multi-path points in the initial path point set, filters the path points, generates an optimized path point set, and passes it to the motion optimization module. The motion optimization module, based on the optimized path point set, obtains the joint positions corresponding to multiple path points and performs path point sequence consistency verification and joint displacement continuity judgment to generate an optimized robot motion path.

Images