A redundant robot arm bagging trajectory generation and pose planning method

Abstract

This invention discloses a method for generating and planning the bagging trajectory of a redundant robotic arm, relating to the field of industrial automation packaging technology. The method involves real-time monitoring of the joint configuration and real-time collision status of the redundant robotic arm on the initial bagging trajectory, dynamically deciding whether to update the initial bagging trajectory based on the monitoring results; based on the decision results of the initial bagging trajectory, preliminary attitude planning is performed using the Jacobi pseudo-inverse null space projection method; the results of the initial bagging trajectory and the preliminary attitude planning are structurally merged, and through trajectory-attitude fusion and collaborative optimization, a final trajectory suitable for bagging operations is generated, solving the problems of low bagging operation accuracy and low operational reliability in dynamic scenarios in existing technologies.

Description

Technical Field

[0001] This invention relates to the field of industrial automated packaging technology, and in particular to a method for generating and planning the bag-packing trajectory of a redundant robotic arm. Background Technology

[0002] In bagging operations within industrial automated packaging, robotic arms are suitable for simple scenarios where the target object is in a fixed position, the surrounding space is open, and the bag opening is stable, such as bagging an open, fixed container. However, in complex scenarios, greater flexibility and environmental adaptability are required, where redundant robotic arms demonstrate their unique advantages. These scenarios often include congested environments around the target object (such as pipelines and instruments surrounding the material cylinder) or situations where the bag opening posture needs dynamic fine-tuning (such as dealing with deformation of soft bag openings or material displacement). Redundant robotic arms can flexibly adjust their posture to bypass obstacles and achieve precise alignment with the target, making them particularly suitable for bagging tasks requiring high precision and operating in complex and variable environments.

[0003] Redundant robotic arms typically refer to robotic arms with more than six degrees of freedom. Their typical structure consists of a series of rotary joints and links connected in series, forming an open-chain kinematic structure. From the base to the end effector, this includes base joints, shoulder joints, elbow joints, wrist joints, and other joints. This series connection ensures that the movement of each joint transmits and affects the position and orientation of all subsequent links and the end effector, thus providing the end effector with more than the minimum number (six) of degrees of freedom required to determine its pose (three positional degrees of freedom and three orientation degrees of freedom) in three-dimensional space. These extra degrees of freedom are called redundant degrees of freedom, which allow the robotic arm to perform infinitely many solutions to its joint configuration when completing a given end effector pose task. Theoretically, this allows for self-motion to optimize internal performance indicators such as obstacle avoidance and energy consumption without affecting the end effector task. Within the task space, an end effector spatial motion trajectory satisfying geometric obstacle avoidance constraints is generated based on environmental perception information. By solving the inverse kinematics, one or more feasible joint angle combinations are obtained, thereby determining the robotic arm's posture sequence.

[0004] In the field of robotic arm trajectory planning and control, existing technologies provide a variety of solutions. For example, Chinese invention patent CN110561426B discloses a path planning method, device and control system for a super-redundant robotic arm, including: after configuring the robotic arm, obtaining arm shape description parameters and planning a desired path with multiple desired nodes; taking the first desired node as the target desired node, obtaining the parameters of the target desired node; obtaining the current parameters and Jacobian matrix of the equivalent arm end through the arm shape description parameters, and obtaining the desired end speed; driving the end of the robotic arm towards the target desired node according to the joint driving parameters; and sequentially traversing the desired nodes to complete the path planning of the robotic arm.

[0005] For example, Chinese invention patent CN111923039B discloses a redundant robotic arm path planning method based on reinforcement learning, which includes: constructing a redundant robotic arm training environment based on a deep reinforcement learning algorithm, and designing a reward function to train the robotic arm.

[0006] However, although the above solutions address some issues in trajectory generation, tracking, or detection, existing posture planning schemes have shortcomings in the coordinated optimization of posture planning with constraints such as obstacle avoidance and joint limits. The dynamic coordination mechanism of bagging process constraints, kinematic constraints, and environmental constraints is poor. In other words, redundant degrees of freedom often exhibit lag in the dynamic adjustment of multiple constraints when dynamically coordinating them, making it difficult to achieve real-time prioritization of different constraints. This leads to planning failures, motion overshoots, or response delays in complex packaging scenarios due to joint motion conflicts and unclear constraint priorities, thereby reducing the overall success rate and adaptability of bagging operations. In complex packaging scenarios, there is also the problem of low bagging operation efficiency due to the underutilization of redundant degrees of freedom of the redundant robotic arm. Summary of the Invention

[0007] To address the technical problems of low accuracy and reliability in bagging operations under dynamic scenarios in existing technologies, this invention provides a method for generating and planning the bagging trajectory of a redundant robotic arm, as follows:

[0008] This invention provides a method for generating and planning the trajectory and attitude of a redundant robotic arm for bagging, comprising the following steps:

[0009] Step 1: Monitor the joint configuration and real-time collision status of the redundant robotic arm on the initial bagging trajectory in real time, and make dynamic decisions on whether to update the initial bagging trajectory based on the monitoring results.

[0010] Step 2: Based on the decision results of the initial bagging trajectory, perform preliminary attitude planning using the Jacobi pseudo-inverse null space projection method;

[0011] Step 3: The initial bagging trajectory and the results of the preliminary attitude planning are structurally merged. Through trajectory-attitude fusion and collaborative optimization, the final trajectory suitable for bagging operations is generated.

[0012] Furthermore, the monitoring of the joint configuration and real-time collision status of the redundant robotic arm on the initial bagging trajectory includes the monitoring of the joint configuration and the monitoring of the real-time collision status after the joint configuration monitoring results are qualified.

[0013] The monitoring process for the joint configuration is as follows:

[0014] Obtain compliance verification parameters corresponding to each sampling point on the initial bagging trajectory of the redundant robotic arm, perform compliance judgment, obtain compliance judgment results, and simultaneously summarize the monitoring results of the joint configuration. The initial bagging trajectory is the reference bagging path for the redundant robotic arm to perform bagging operations, and the joint configuration represents the correspondence between the end effector pose and the joint spatial angle distribution of the redundant robotic arm at each sampling time on the initial bagging trajectory.

[0015] The compliance determination includes joint limit compliance determination, unusual configuration compliance determination, and operability measurement compliance determination;

[0016] The compliance assessment results include joint limit compliance assessment results, singular configuration compliance assessment results, and operability measurement compliance assessment results;

[0017] The joint limit compliance determination is as follows: within a preset time period, the rotation angle of each joint of the redundant robotic arm is compared with the preset rotation angle range. If the rotation angle of any joint is not within the preset rotation angle range, it indicates that the joint limit status of the redundant robotic arm is unqualified. If the actual rotation angle of all joints is within their respective preset rotation angle ranges, it indicates that the joint limit status of the redundant robotic arm is qualified.

[0018] The singular configuration compliance determination is as follows: Singular value decomposition is performed on the Jacobian matrix corresponding to the current joint configuration of the redundant robot arm, and the minimum singular value of the Jacobian matrix is ​​extracted. If the obtained minimum singular value is less than the preset singular avoidance threshold, it indicates that the singular configuration state of the redundant robot arm is unqualified; if the obtained minimum singular value is greater than or equal to the preset singular avoidance threshold, it indicates that the singular configuration state of the redundant robot arm is qualified.

[0019] The compliance determination of the operability measurement is as follows: Calculate the determinant of the product of the Jacobian matrix and its transpose matrix under the current joint configuration of the redundant robot arm, and take the square root of the determinant to obtain the operability measurement value. If the obtained operability measurement value is lower than the preset minimum operability threshold, it indicates that the operability measurement status of the redundant robot arm is unqualified. If the obtained operability measurement value is greater than or equal to the preset minimum operability threshold, it indicates that the operability measurement status of the redundant robot arm is qualified.

[0020] Furthermore, the monitoring results of the summarized joint configuration are specifically as follows:

[0021] The joint configuration monitoring results are deemed qualified only if all compliance judgment results are qualified, and real-time collision status monitoring is carried out.

[0022] Conversely, if the joint configuration monitoring results are not satisfactory, zero-space vector iterative optimization is performed to redistribute redundant degrees of freedom. After the redistribution of redundant degrees of freedom is completed, real-time collision status monitoring is performed.

[0023] Furthermore, the step of performing null space vector iterative optimization to reallocate redundant degrees of freedom specifically involves:

[0024] For joint configurations that are determined to be unqualified by joint configuration monitoring results, the set of gradient vector deviation rates corresponding to the rotation angle of each joint under the joint configuration is obtained and projected into the zero space of the Jacobian matrix of the corresponding joint configuration to obtain the zero space optimization direction vector.

[0025] By optimizing the direction vector in null space and combining it with joint constraint conditions, the redundant degrees of freedom of the current joint configuration are iteratively corrected. By gradually adjusting the angles of each joint, the redundant degrees of freedom are redistributed, and the collision status is monitored in real time.

[0026] Furthermore, the real-time collision status monitoring specifically includes:

[0027] The joint configurations that meet the monitoring results are recorded as qualified joint configurations. The link distances of the redundant robotic arm at each sampling point on the initial bagging trajectory under the qualified joint configurations are obtained, and the minimum link distance is selected from them.

[0028] When the minimum distance between the links is within the first distance range, it indicates that the corresponding risk level is the first risk level, and a collision risk warning is triggered based on the current joint configuration.

[0029] When the minimum distance of the link is within the second distance range, it indicates that the corresponding risk level is the second risk level. The initial bagging trajectory is updated based on the current joint configuration, and preliminary attitude planning is carried out after the update is qualified.

[0030] When the minimum distance of the link is within the third distance range, it indicates that the corresponding risk level is the third risk level, and preliminary attitude planning is carried out directly.

[0031] The degree of risk decreases sequentially from the first risk level, the second risk level, to the third risk level.

[0032] Furthermore, the preliminary attitude planning process is as follows:

[0033] Based on the joint configuration within the third distance interval, the actual rotation angle of each joint under the corresponding joint configuration is obtained and used as the initial target angle parameter;

[0034] The target angle parameters of the joint configurations within the third distance interval are sorted and integrated according to the sampling time sequence of the initial bagging trajectory, and associated with the initial target angle parameters of each sampling point to form a preliminary joint configuration sequence.

[0035] By setting smoothing constraints based on the kinematic characteristics of the redundant robotic arm, the initial joint configuration sequence is smoothed and optimized to obtain a continuous optimized joint configuration sequence.

[0036] Furthermore, the smoothing optimization process is performed to obtain a continuous optimized joint configuration sequence. The specific process is as follows:

[0037] In the preliminary joint configuration sequence, the rate of change of joint rotation angle between the current sampling point and the joint configuration corresponding to the previous sampling point is calculated. The rate of change of joint rotation angle is used to reflect the speed of change of joint rotation angle between adjacent sampling points. If there is no previous sampling point, the rate of change of joint rotation angle is defined as 0.

[0038] Summarize the rate of change of joint rotation angle between all adjacent sampling points in the preliminary joint configuration sequence to form a complete set of rates of change.

[0039] If there is a joint configuration in the set of rates of change that has a rate of change of joint rotation angle greater than the preset rate of change of joint rotation angle, it indicates that there is a sudden change in the initial posture sequence and it is removed.

[0040] If there is no joint configuration in the set of rates of change that has a rate of change of joint rotation angle greater than the preset rate of change of joint rotation angle, it indicates that the current preliminary posture sequence is in a stable state, and the redundant robotic arm is prompted to maintain the current posture. At the same time, the corresponding joint configuration is recorded as a qualified joint configuration and structured merging is performed.

[0041] Furthermore, the structured merging specifically refers to:

[0042] The end effector poses corresponding to qualified joint configurations in the initial joint configuration sequence are mapped to the time axis of the end position sequence obtained by dynamic optimization trajectory discretization. Each timestamp on the time axis is simultaneously associated with the corresponding end effector position and end effector pose to obtain a time-synchronized trajectory pose sequence.

[0043] Calculate the spatial angle between the direction of the end effector axis and the direction of the trajectory tangent at each time stamp in the trajectory pose sequence, and use this spatial angle as the attitude-trajectory matching degree;

[0044] If the spatial angle is not greater than the preset spatial angle, that is, the attitude-trajectory matching degree is greater than the preset attitude-trajectory matching degree, then the fusion and collaborative optimization are directly performed based on the trajectory pose sequence.

[0045] Conversely, the attitude-trajectory matching deviation rate is obtained, and within the null space of the qualified joint configuration corresponding to the current timestamp, the spatial angle is reduced as the optimization objective to perform iterative adjustment, resulting in a new joint configuration and the corresponding end effector attitude. Then, the trajectory pose sequence after iterative adjustment is fused and collaboratively optimized.

[0046] Furthermore, when the change in rotation angle of any joint in the trajectory pose sequence exceeds the preset change in joint rotation angle, the specific steps of the fusion and collaborative optimization are as follows:

[0047] The joint configuration and target joint corresponding to the change in rotation angle exceeding the preset change in joint rotation angle are obtained. The time interval corresponding to the change in joint rotation angle exceeding the preset change in joint rotation angle is taken as the abrupt change time interval. The average angular velocity of the joint is calculated based on the target joint rotation angle value at the beginning and end of the interval.

[0048] If the average angular velocity of a joint exceeds the preset average angular velocity of a joint, the corresponding joint is determined to be in a state of velocity incoordination, and the movement time of the corresponding joint is extended by inserting at least one intermediate time point within the abrupt change time interval.

[0049] Conversely, the output trajectory pose sequence is used as the final bagging trajectory.

[0050] Furthermore, when there is no rotation angle change exceeding the preset joint rotation angle change in the trajectory pose sequence, the specific steps of the fusion and collaborative optimization are as follows:

[0051] Calculate the absolute value of the deviation between the rotation angle of each joint and the median of the limit in the trajectory pose sequence, and use it as the joint deviation value;

[0052] If any joint has a joint deviation value that exceeds the corresponding preset joint deviation value, then perform redundancy enhancement posture optimization; otherwise, directly output the trajectory pose sequence as the final bagging trajectory.

[0053] The redundancy enhancement posture optimization specifically involves: projecting the gradient of the target joint rotation angle moving towards the limit median direction into the null space, optimizing to reduce the joint deviation value, obtaining the optimized trajectory pose sequence, and outputting the optimized trajectory pose sequence as the final bagging trajectory.

[0054] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0055] (1) This invention establishes a multi-dimensional compliance judgment system based on joint limits, singular configurations, and operability measurements to achieve comprehensive real-time monitoring of the joint configuration of redundant robotic arms on the initial bagging trajectory. This mechanism not only focuses on whether the joint angles exceed mechanical limits, but also evaluates the kinematic performance and stability of the redundant robotic arm under the current configuration through the singular value decomposition of the Jacobian matrix and operability measurements. This multi-constraint collaborative monitoring method, by combining joint configuration monitoring with real-time collision status monitoring, can identify potential risks before trajectory execution and dynamically decide whether to update the trajectory or redistribute the degrees of freedom. This risk identification and response mechanism enables the redundant robotic arm to have stronger adaptive capabilities in complex and dynamic packaging environments, and can cope with uncertainties such as sudden obstacles and bag opening deformation, thereby improving the continuity and success rate of bagging operations.

[0056] (2) This invention combines Jacobi pseudo-inverse solving with null space projection technology for the initial allocation and iterative optimization of redundant degrees of freedom. In the initial posture planning stage, Jacobi pseudo-inverse is used to ensure that the end effector moves along the predetermined trajectory, while the self-motion corresponding to the redundant degrees of freedom is projected to the null space. Thus, without affecting the end task, the internal performance indicators of the redundant manipulator are optimized, such as moving away from joint limits and avoiding singular configurations. When the joint configuration is found to be unqualified, the gradient vector deviation rate is calculated and projected to the null space of the current configuration Jacobi matrix to obtain the optimized direction vector, and then the joint angle is iteratively corrected. This local optimization in the null space realizes the efficient and accurate redistribution of redundant degrees of freedom and avoids the computational overhead caused by global replanning. It enables the manipulator to complete the same end task with a better self posture, improving motion efficiency and energy utilization.

[0057] (3) This invention proposes a graded collision risk monitoring strategy. By calculating the minimum distance between each link of the robotic arm at the trajectory sampling point and mapping it to a preset distance range, the collision risk can be assessed in a refined and quantitative manner. For the first risk range, an early warning is triggered immediately. For the second risk range, the initial trajectory update process is triggered. Only the third risk range directly enters the subsequent attitude planning. This graded response mechanism avoids the conservative or insufficient response problems caused by a one-size-fits-all obstacle avoidance strategy. For medium-risk scenarios, the initial bagging trajectory is actively updated to generate a safer alternative path, and the operation continues after the update is qualified. This dynamic trajectory update capability enables the robotic arm to maintain the continuity of operation to the maximum extent while ensuring safety. It is suitable for crowded scenarios with dense pipelines and instruments around the target object and narrow space, and achieves a high level of balance between safety and operation efficiency.

[0058] (4) The fusion and collaborative optimization of the present invention comprehensively reviews and optimizes the generated trajectory pose sequence; in order to deal with the speed incoordination that may be caused by sudden changes in joint angle, the motion is re-planned by inserting intermediate points in the sudden change interval and extending the motion time, so as to ensure smooth motion and not exceed the joint speed capability; in order to deal with the situation where the joint angle is too close to the limit position, the posture optimization is improved by redundancy, and the joint angle is adjusted towards the limit median to increase the margin of subsequent motion; it is achieved by zero-space projection technology, and all of them are based on ensuring the completion of the end task; so that the final output bagging trajectory not only meets the basic requirements of obstacle avoidance and positioning, but also achieves comprehensive optimization in terms of motion smoothness, joint space comfort, and future adjustment margin, so that the redundant robotic arm exhibits stronger overall stability, adaptability and durability when dealing with complex bagging tasks with long time, high frequency and changing environment. Attached Figure Description

[0059] 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.

[0060] Figure 1 This is a flowchart of a redundant robotic arm bag-handling trajectory generation and attitude planning method according to the present invention;

[0061] Figure 2 This is a flowchart showing the correspondence between the joint configuration and real-time collision state monitoring of the present invention;

[0062] Figure 3 This is a flowchart corresponding to the preliminary attitude planning of the present invention;

[0063] Figure 4 This is a flowchart corresponding to the structured merging of the present invention. Detailed Implementation

[0064] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0065] 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.

[0066] 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.

[0067] Example 1 provides a method for generating and planning the trajectory and attitude of a redundant robotic arm for bagging, such as... Figure 1 The flowchart shown is a method for generating and planning the bagging trajectory of a redundant robotic arm. The processing flow of this method can include the following steps: Step 1: Real-time monitoring of the joint configuration and real-time collision status of the redundant robotic arm on the initial bagging trajectory, and dynamic decision on whether to update the initial bagging trajectory based on the monitoring results; Step 2: Based on the decision results of the initial bagging trajectory, preliminary attitude planning is performed using the Jacobi pseudo-inverse null space projection method; Step 3: The initial bagging trajectory and the results of the preliminary attitude planning are structurally merged, and the final trajectory suitable for bagging operations is generated through trajectory-attitude fusion and collaborative optimization.

[0068] like Figure 2 The diagram shown is a flowchart illustrating the correspondence between joint configuration and real-time collision state monitoring provided in an embodiment of the present invention. Specifically, it includes joint configuration monitoring and real-time collision state monitoring performed after the joint configuration monitoring results are deemed satisfactory. The joint configuration monitoring process is as follows:

[0069] The compliance verification parameters corresponding to each sampling point on the initial bagging trajectory of the redundant robotic arm are acquired, including rotation angle, minimum singular value, and operability metric value. Compliance judgment is then performed to obtain the compliance judgment result. The monitoring results of the joint configuration are summarized simultaneously. The initial bagging trajectory is a reference bagging path pre-set by the preset personnel for the redundant robotic arm to perform bagging operations. The joint configuration represents the correspondence between the end effector pose and the spatial angle distribution of the joints at each sampling moment of the trajectory. This correspondence represents a unique mapping based on forward kinematics. The sampling point refers to the spatial position point pre-selected by equal interpolation on the initial bagging trajectory. The time corresponding to each sampling point is the sampling time. The compliance judgment includes joint limit compliance judgment, singular configuration compliance judgment, and operability metric compliance judgment. The compliance judgment result includes the joint limit compliance judgment result, singular configuration compliance judgment result, and operability metric compliance judgment result.

[0070] The joint limit compliance determination is as follows: within a preset time period, the rotation angle of each joint of the redundant robotic arm is compared with a preset rotation angle range. If the rotation angle of any joint is not within the preset rotation angle range, it indicates that the joint limit status of the redundant robotic arm is unqualified. If the actual rotation angle of all joints is within their respective preset rotation angle ranges, it indicates that the joint limit status of the redundant robotic arm is qualified. The joint limit status refers to the degree to which the current rotation angle of each rotating joint of the redundant robotic arm is close to its physical motion limit (hard limit). The current angle of all joints is within a preset rotation angle range. The preset rotation angle range is the angle working range pre-set for each rotating joint of the redundant robotic arm based on its physical mechanical limit and safety margin. The preset time period is pre-set by the preset personnel according to the bagging task requirements of the redundant robotic arm under different conditions, usually 5 seconds.

[0071] The singular configuration compliance determination is as follows: For the Jacobian matrix corresponding to the current joint configuration of the redundant robotic arm, a standard singular value decomposition mathematical operation is performed on the Jacobian matrix, decomposing it into the product of three specific matrices. In the diagonal matrix generated by this decomposition, the singular values ​​on the diagonal are arranged from largest to smallest. The smallest value in this set of singular values ​​is selected as the minimum singular value. If the obtained minimum singular value is less than the preset singular avoidance threshold, it indicates that the singular configuration state of the redundant robotic arm is unqualified. If the obtained minimum singular value is greater than or equal to the preset singular avoidance threshold, it indicates that the singular configuration state of the redundant robotic arm is qualified. The preset singular avoidance threshold is pre-set based on the stability requirements of the robotic arm's motion control. When the minimum singular value of the Jacobian matrix is ​​lower than this threshold, it means that the robotic arm is in or close to a kinematically singular configuration.

[0072] The compliance determination of the operability measurement is as follows: Calculate the determinant of the product of the Jacobian matrix and its transpose matrix under the current joint configuration of the redundant robotic arm, and take the square root of the determinant to obtain the operability measurement value. If the obtained operability measurement value is lower than the preset minimum operability threshold, it indicates that the operability measurement status of the redundant robotic arm is unqualified. If the obtained operability measurement value is greater than or equal to the preset minimum operability threshold, it indicates that the operability measurement status of the redundant robotic arm is qualified. The preset minimum operability threshold is preset by the preset personnel based on the minimum requirement of the robotic arm end-effector flexibility for the specific bagging task.

[0073] The rotation angle of each joint is obtained by reading the feedback from the high-precision encoder installed at each joint. The Jacobian matrix is ​​calculated online based on the kinematics of the robotic arm (such as by using the DH parameter method) and the real-time joint angle vector. Singular value decomposition and determinant calculation are standard numerical operations for this Jacobian matrix.

[0074] The monitoring result of the joint configuration is determined to be qualified and the real-time collision state is monitored if and only if all compliance determination results are qualified, that is, the redundant robotic arm joint limit state is qualified, the redundant robotic arm singular configuration state is qualified, and the redundant robotic arm manipulability metric state is qualified; otherwise, the monitoring result of the joint configuration is determined to be unqualified, and null space vector iterative optimization is performed to reallocate redundant degrees of freedom. After the allocation of redundant degrees of freedom is completed, the real-time collision state is monitored, specifically as follows:

[0075] For the joint configuration determined to have an unqualified monitoring result of the joint configuration, obtain the set of gradient vector deviation rates corresponding to the rotation angles of each joint in this joint configuration and project it into the null space of the Jacobian matrix corresponding to the joint configuration, that is, multiply the gradient vector deviation rates in the set of gradient vector deviation rates by this projection matrix. After this mathematical projection operation, the new vector obtained is the null space optimization direction vector; the specific process of obtaining the set of gradient vector deviation rates is as follows: According to the specific unqualified type (joint limit, singular configuration, or insufficient manipulability), a penalty function constructed based on the historical joint angle vector, whose input is the current joint angle vector and output is the corresponding gradient vector. Among them, the joint limit setting corresponding joint limit penalty term encourages the joint to stay away from the limit, the singularity avoidance corresponding singularity avoidance penalty term encourages the configuration to stay away from singularities, and the manipulability improvement penalty term is used to encourage increased flexibility. By using analytical differentiation or numerical differentiation methods, calculate the gradient vector of the differentiable penalty function with respect to the joint angle vector. This gradient points in the direction that makes the performance index increase the fastest, so the negative gradient direction is the performance improvement direction.

[0076] Calculate the pseudoinverse of the Jacobian matrix based on the current configuration to obtain the null space projection matrix, and project the performance improvement direction into this null space to obtain the pure self-motion adjustment direction that does not change the end pose, which is the null space optimization direction vector.

[0077] Iteratively correct the redundant degrees of freedom of the current joint configuration through the null space optimization direction vector combined with the joint limit constraint conditions. By gradually adjusting the angles of each joint, complete the reallocation of redundant degrees of freedom and monitor the real-time collision state.

[0078] The joint limit constraint condition is an inherent physical property of the redundant robotic arm, referring to the minimum and maximum angles allowed for each rotating joint. These data are obtained from the factory settings of the robotic arm; iterative correction is a cyclic process that combines gradient descent and constraint handling. Based on the null space optimization direction vector under the current joint configuration, iterate along the optimization direction. Starting from the current joint configuration, according to the calculated null space optimization direction vector, update the joint angle vector along this direction with an adaptive step size in each iteration.

[0079] Specifically, the real-time collision state monitoring process is as follows: the joint configuration corresponding to the qualified joint configuration is recorded as the qualified joint configuration. The link distances of the redundant robotic arm at each sampling point on the initial bagging trajectory are obtained through the Euclidean distance between the link and the obstacle. That is, during the movement with the initial bagging trajectory, the Euclidean distances between the link and the obstacle are obtained through sampling, and the minimum link distance is selected from them.

[0080] When the minimum distance of the link is within the first distance range, it indicates that the corresponding risk level is the first risk level. Based on the current joint configuration, a collision risk warning is triggered, such as pausing the current bagging action and waiting for confirmation or intervention from the preset personnel.

[0081] When the minimum distance of the link is within the second distance range, it indicates that the corresponding risk level is the second risk level. The initial bagging trajectory is updated based on the current joint configuration, and preliminary attitude planning is performed after the update is qualified.

[0082] When the minimum distance of the link is within the third distance range, it indicates that the corresponding risk level is the third risk level, and preliminary attitude planning is carried out directly; the distance risk level of the first risk level, the second risk level and the third risk level decreases in sequence.

[0083] When the minimum distance between the links is detected to be within the second distance range, the movement of the robotic arm along the current initial bagging trajectory is immediately paused, and the joint configuration, end effector pose, and detected obstacle information at the current moment are recorded. While keeping the current position and attitude of the end effector unchanged, local null space optimization is initiated. This optimization aims to increase the distance between all links and obstacles under the current joint configuration. Iterative calculations are performed within the null space of the current joint configuration. After verification, the updated initial bagging trajectory is output, and preliminary attitude planning is performed.

[0084] like Figure 3The diagram shows a flowchart of the preliminary attitude planning provided in this embodiment of the invention. The specific process of preliminary attitude planning is as follows: Taking the joint configuration within the third distance interval as a reference, the actual rotation angle of each joint under the corresponding joint configuration is obtained and used as the initial target angle parameter; the target angle parameters corresponding to the joint configuration within the third distance interval of each sampling point are sorted and integrated according to the sampling time sequence of the initial bagging trajectory, and all sampling points within the safe distance (third distance interval) are selected. Their timestamps and corresponding joint rotation angles are extracted, and these data points are arranged in ascending order of timestamps. The sorted data is organized into an ordered list and associated with the initial target angle parameters of each sampling point to form a preliminary joint configuration sequence; combined with the kinematic characteristics of the redundant robotic arm (including joint velocity limit, acceleration limit and joint coupling constraint), smoothing constraint conditions are set, and the preliminary joint configuration sequence is smoothed and optimized to obtain a continuous and stable optimized joint configuration sequence, specifically:

[0085] In the preliminary joint configuration sequence, the rate of change of joint rotation angle between the current sampling point and the joint configuration corresponding to the previous sampling point is calculated. The rate of change of joint rotation angle is obtained by proportionalizing the difference between the current joint rotation angle and the joint rotation angle of the previous sampling point as the numerator and the current joint rotation angle as the denominator. The rate of change of joint rotation angle is used to reflect the speed of change of joint rotation angle between adjacent sampling points and is the core indicator for judging whether there is abrupt change in the joint configuration sequence. If there is no previous sampling point, that is, the current sampling point is the first sampling point of the initial bagging trajectory, the rate of change of joint rotation angle is defined as 0, indicating that the joint rotation angle of the current sampling point has no change.

[0086] In the preliminary joint configuration sequence, the rate of change of joint rotation angle between all adjacent sampling points is summarized. Starting from the first sampling point in the sequence, each pair of adjacent sampling points is visited in turn. The preliminary joint configuration sequence is traversed. For the currently visited pair of adjacent points, the rate of change of joint rotation angle is determined and stored in a preset list or array in turn to form a complete set of rates of change.

[0087] If there is a joint rotation angle change rate in the set of change rates that is greater than the preset joint rotation angle change rate, and the corresponding joint configuration is there, it indicates that there is a sudden change in the initial posture sequence, and it is removed. The preset joint rotation angle change rate is determined by offline dynamic simulation. If there is no joint rotation angle change rate in the set of change rates that is greater than the preset joint rotation angle change rate, it indicates that the current initial posture sequence is in a stable state, and the redundant robotic arm is prompted to maintain the current posture. At the same time, the corresponding joint configuration is recorded as a qualified joint configuration and merged in a structured manner.

[0088] like Figure 4The flowchart shown is a flowchart of the structured merging process provided in this embodiment of the invention. The specific steps of the structured merging process are as follows: For the end effector postures corresponding to all qualified joint configurations in the preliminary joint configuration sequence, the end position sequence is discretized and aligned with the time axis based on a common time reference. By matching or interpolation, each posture is assigned a timestamp corresponding to the end position sequence so that there is a unique end position and posture at the same time. Finally, the position vector and posture associated with each timestamp are arranged in chronological order to form a trajectory pose sequence. The end effector posture is the spatial form of the end effector at the current timestamp.

[0089] The spatial angle between the end effector axis direction and the trajectory tangent direction at each time stamp in the trajectory pose sequence is calculated, and this spatial angle is used as the attitude-trajectory matching degree. If the attitude-trajectory matching degree is greater than the preset attitude-trajectory matching degree, fusion and collaborative optimization are directly performed based on the trajectory pose sequence. Otherwise, the attitude-trajectory matching degree deviation rate is obtained. Within the null space of the qualified joint configuration corresponding to the current time stamp, iterative adjustments are made with the reduction of the spatial angle as the optimization objective to obtain a new joint configuration and the corresponding end effector attitude. Then, fusion and collaborative optimization are performed based on the iteratively adjusted trajectory pose sequence. The matching degree deviation rate is obtained by proportionally processing the obtained attitude-trajectory matching degree as the numerator and the preset attitude-trajectory matching degree as the denominator. The preset attitude-trajectory matching degree is represented by the sum and average of the historical attitude-trajectory matching degrees, and the preset spatial angle is represented by the sum and average of the historical spatial angles.

[0090] Iterative adjustments are performed in the null space of the joint configuration at the corresponding timestamp. The core is to adjust the joint angle to change the end effector's attitude without changing the end effector's position, thereby reducing the angle between its axis and the trajectory tangent. Specifically, with this angle as the optimization target, the gradient of the angle with respect to the joint angle vector is calculated. Using the kinematic chain rule, the Jacobian matrix of the gradient of the end effector's axis direction with respect to the joint angle vector and its null space projection operator are calculated. The gradient is projected onto the null space to obtain the optimized direction without changing the end effector's position. The joint angle is then iteratively updated, and the optimized joint configuration and its corresponding new end effector attitude are output.

[0091] When there is no change in rotation angle in the trajectory pose sequence that exceeds the preset change in joint rotation angle, the specific steps of fusion and collaborative optimization are as follows: calculate the absolute value of the deviation between the rotation angle of each joint in the trajectory pose sequence and the limit median, and use it as the joint deviation value. The limit median is the physical limit median, which is based on an inherent physical parameter of the redundant robotic arm, namely the arithmetic mean of the minimum and maximum rotation angles allowed for each rotation joint.

[0092] If any joint deviation exceeds the corresponding preset joint deviation value, redundancy enhancement posture optimization is performed; otherwise, the trajectory pose sequence is directly output as the final bagging trajectory. The preset joint deviation value is a quantitative standard for judging whether a joint is too close to its physical limit and needs to be optimized to improve redundancy. Based on the physical workspace of the robotic arm, the total range of motion of each joint is calculated and a maximum allowable deviation proportional to this range is set as the preset joint deviation value. For example, 20% of the total range is taken as the preset joint deviation value.

[0093] Redundancy enhancement and posture optimization are specifically achieved by: aiming to reduce joint deviation, projecting the gradient of the target joint rotation angle moving towards the median of the limit to the null space to obtain the optimized trajectory pose sequence, and using the optimized trajectory pose sequence as the final bagging trajectory output; the optimization process first determines the optimization target and direction, and without changing the current position and posture of the end effector (i.e. without affecting the predetermined bagging task), by adjusting the joint angles inside the robotic arm, joints that are too close to the physical limit can regress to their optimized trajectory pose sequence, thereby improving the redundancy of the entire robotic arm in response to subsequent uncertainties.

[0094] In Example 1, through online monitoring of joint configuration and real-time collision status assessment, the robotic arm can autonomously identify and avoid risks such as joint limits, unusual configurations, and insufficient operational flexibility during movement, thereby reducing the probability of sudden stops, vibrations, or loss of control. In complex and unstructured actual operation scenarios, such as the presence of dynamic obstacles or minor disturbances at the bag opening position, this method improves the environmental adaptability of the redundant robotic arm. Through real-time collision monitoring and graded early warning mechanisms, different levels of risk can be distinguished and corresponding strategies can be adopted. This enables the robotic arm to maintain operational continuity while ensuring safety, reducing unnecessary interruptions and improving overall operational efficiency.

[0095] Furthermore, by introducing collaborative optimization based on zero-space projection, the end-effector trajectory tracking accuracy and robotic arm motion are unified. The generated final bagging trajectory not only ensures that the end effector can accurately and stably complete the preset bagging path and posture planning, but also makes the movement of each joint of the robotic arm smoother and more coordinated through optimization. This effectively reduces the impact, vibration and noise during the movement, making the bagging action smooth and natural, and improving the success rate of handling soft bag openings or fragile materials.

[0096] Before fusion and collaborative optimization, joint rotation angles may exceed the preset range in the trajectory pose sequence. That is, when there is no rotation angle change in the trajectory pose sequence that exceeds the preset joint rotation angle change, the fusion and collaborative optimization process corresponding to the scenario in Example 2 is provided, and its specific steps are as follows:

[0097] In Example 2, when there is no rotation angle change exceeding the preset joint rotation angle change in the trajectory pose sequence, the joint configuration and target joint corresponding to the rotation angle change exceeding the preset joint rotation angle change are obtained, and their respective time intervals are used as abrupt change time intervals. The average joint angular velocity is obtained by arithmetic averaging the joint angle values ​​at the start and end points of the interval. The joint rotation angle change is obtained by calculating the absolute value of the rotation angle difference of the same joint in adjacent sampling periods. The preset joint rotation angle change is preset based on the physical limits of the joint and the requirements for motion stability.

[0098] If the average angular velocity of a joint exceeds the preset average angular velocity, it is determined to be a speed incoordination. In this case, at least one intermediate time point is inserted within the abrupt change time interval to extend the motion time. That is, the total time to be extended is calculated based on the duration of the abrupt change time interval to meet the angular velocity requirement: Total extended time = Total change in joint rotation angle within the abrupt change time interval / Average angular velocity of the joint - Duration of the abrupt change time interval. A preset number of intermediate time points are inserted on average within the extended total time to re-plan the joint motion within this interval. The preset average angular velocity of the joint is a limit set based on the different speed requirements of each stage of the bagging operation (such as rapid approach and fine alignment). Otherwise, the output trajectory pose sequence is used as the final bagging trajectory.

[0099] In Example 2, through online optimization of motion curves, local velocity peaks can be actively identified and smoothed, transforming abrupt motions that cause robotic arm tremors into smooth, continuous motion transitions. This not only improves the stability and smoothness of trajectory execution and reduces the impact of motion shocks on the mechanical body and transmission components, but also reduces end-effector tremors caused by sudden velocity changes, thereby ensuring the stability and accuracy of bagging operations. At the same time, this optimization is completed while maintaining the continuity of operations, without interrupting the task, thus improving the efficiency and adaptability of bagging operations.

[0100] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the flow or function according to the embodiments of the present invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0101] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0102] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0103] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0104] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0105] In the embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0106] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0107] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0108] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0109] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for generating and planning the trajectory and attitude of a redundant robotic arm for bagging, characterized in that, Includes the following steps: Step 1: Monitor the joint configuration and real-time collision status of the redundant robotic arm on the initial bagging trajectory in real time, and dynamically decide whether to update the initial bagging trajectory based on the monitoring results. Step 2: Based on the decision results of the initial bagging trajectory, perform preliminary attitude planning using the Jacobi pseudo-inverse null space projection method; Step 3: The initial bagging trajectory and the results of the preliminary attitude planning are structurally merged. Through trajectory-attitude fusion and collaborative optimization, the final trajectory suitable for bagging operation is generated. The monitoring of the joint configuration and real-time collision status of the redundant robotic arm on the initial bagging trajectory includes the monitoring of the joint configuration and the monitoring of the real-time collision status after the joint configuration monitoring results are qualified. The monitoring process for the joint configuration is as follows: Obtain compliance verification parameters corresponding to each sampling point on the initial bagging trajectory of the redundant robotic arm, perform compliance judgment, obtain compliance judgment results, and simultaneously summarize the monitoring results of the joint configuration. The initial bagging trajectory is the reference bagging path for the redundant robotic arm to perform bagging operations, and the joint configuration represents the correspondence between the end effector pose and the joint spatial angle distribution of the redundant robotic arm at each sampling time on the initial bagging trajectory. The compliance determination includes joint limit compliance determination, unusual configuration compliance determination, and operability measurement compliance determination; The compliance assessment results include joint limit compliance assessment results, unusual configuration compliance assessment results, and operability measurement compliance assessment results.

2. The redundant robotic arm bagging trajectory generation and attitude planning method as described in claim 1, characterized in that, The joint limit compliance determination is as follows: within a preset time period, the rotation angle of each joint of the redundant robotic arm is compared with the preset rotation angle range. If the rotation angle of any joint is not within the preset rotation angle range, it indicates that the joint limit status of the redundant robotic arm is unqualified. If the actual rotation angle of all joints is within their respective preset rotation angle ranges, it indicates that the joint limit status of the redundant robotic arm is qualified. The singular configuration compliance determination is as follows: Singular value decomposition is performed on the Jacobian matrix corresponding to the current joint configuration of the redundant robot arm, and the minimum singular value of the Jacobian matrix is ​​extracted. If the obtained minimum singular value is less than the preset singular avoidance threshold, it indicates that the singular configuration state of the redundant robot arm is unqualified; if the obtained minimum singular value is greater than or equal to the preset singular avoidance threshold, it indicates that the singular configuration state of the redundant robot arm is qualified. The compliance determination of the operability measurement is as follows: Calculate the determinant of the product of the Jacobian matrix and its transpose matrix under the current joint configuration of the redundant robot arm, and take the square root of the determinant to obtain the operability measurement value. If the obtained operability measurement value is lower than the preset minimum operability threshold, it indicates that the operability measurement status of the redundant robot arm is unqualified. If the obtained operability measurement value is greater than or equal to the preset minimum operability threshold, it indicates that the operability measurement status of the redundant robot arm is qualified.

3. The redundant robotic arm bagging trajectory generation and attitude planning method as described in claim 2, characterized in that, The monitoring results of the summarized joint configuration are as follows: The joint configuration monitoring results are deemed qualified only if all compliance judgment results are qualified, and real-time collision status monitoring is carried out. Conversely, if the joint configuration monitoring results are not satisfactory, zero-space vector iterative optimization is performed to redistribute redundant degrees of freedom. After the redistribution of redundant degrees of freedom is completed, real-time collision status monitoring is performed.

4. The redundant robotic arm bagging trajectory generation and attitude planning method as described in claim 3, characterized in that, The step of performing zero-space vector iterative optimization to reallocate redundant degrees of freedom specifically involves: For joint configurations that are determined to be unqualified by joint configuration monitoring results, the set of gradient vector deviation rates corresponding to the rotation angle of each joint under the joint configuration is obtained and projected into the zero space of the Jacobian matrix of the corresponding joint configuration to obtain the zero space optimization direction vector. By optimizing the direction vector in null space and combining it with joint constraint conditions, the redundant degrees of freedom of the current joint configuration are iteratively corrected. By gradually adjusting the angles of each joint, the redundant degrees of freedom are redistributed, and the collision status is monitored in real time.

5. The redundant robotic arm bagging trajectory generation and attitude planning method as described in claim 4, characterized in that, The real-time collision status monitoring specifically includes: The joint configurations that meet the monitoring results are recorded as qualified joint configurations. The link distances of the redundant robotic arm at each sampling point on the initial bagging trajectory under the qualified joint configurations are obtained, and the minimum link distance is selected from them. When the minimum distance between the links is within the first distance range, it indicates that the corresponding risk level is the first risk level, and a collision risk warning is triggered based on the current joint configuration. When the minimum distance of the link is within the second distance range, it indicates that the corresponding risk level is the second risk level. The initial bagging trajectory is updated based on the current joint configuration, and preliminary attitude planning is carried out after the update is qualified. When the minimum distance of the link is within the third distance range, it indicates that the corresponding risk level is the third risk level, and preliminary attitude planning is carried out directly. The degree of risk decreases sequentially from the first risk level, the second risk level, to the third risk level.

6. The redundant robotic arm bagging trajectory generation and attitude planning method as described in claim 5, characterized in that, The preliminary attitude planning process is as follows: Based on the joint configuration within the third distance interval, the actual rotation angle of each joint under the corresponding joint configuration is obtained and used as the initial target angle parameter; The target angle parameters of the joint configurations within the third distance interval are sorted and integrated according to the sampling time sequence of the initial bagging trajectory, and associated with the initial target angle parameters of each sampling point to form a preliminary joint configuration sequence. By setting smoothing constraints based on the kinematic characteristics of the redundant robotic arm, the initial joint configuration sequence is smoothed and optimized to obtain a continuous optimized joint configuration sequence.

7. The redundant robotic arm bagging trajectory generation and attitude planning method as described in claim 6, characterized in that, The smoothing optimization process is performed to obtain a continuous optimized joint configuration sequence. The specific process is as follows: In the preliminary joint configuration sequence, the rate of change of joint rotation angle between the current sampling point and the joint configuration corresponding to the previous sampling point is calculated. The rate of change of joint rotation angle is used to reflect the speed of change of joint rotation angle between adjacent sampling points. If there is no previous sampling point, the rate of change of joint rotation angle is defined as 0. Summarize the rate of change of joint rotation angle between all adjacent sampling points in the preliminary joint configuration sequence to form a complete set of rates of change. If there is a joint configuration in the set of rates of change that has a rate of change of joint rotation angle greater than the preset rate of change of joint rotation angle, it indicates that there is a sudden change in the initial posture sequence and it is removed. If there is no joint configuration in the set of rates of change that has a rate of change of joint rotation angle greater than the preset rate of change of joint rotation angle, it indicates that the current preliminary posture sequence is in a stable state, and the redundant robotic arm is prompted to maintain the current posture. At the same time, the corresponding joint configuration is recorded as a qualified joint configuration and structured merging is performed.

8. The redundant robotic arm bagging trajectory generation and attitude planning method as described in claim 7, characterized in that, The structured merging is specifically as follows: The end effector poses corresponding to qualified joint configurations in the initial joint configuration sequence are mapped to the time axis of the end position sequence obtained by dynamic optimization trajectory discretization. Each timestamp on the time axis is simultaneously associated with the corresponding end effector position and end effector pose to obtain a time-synchronized trajectory pose sequence. Calculate the spatial angle between the direction of the end effector axis and the direction of the trajectory tangent at each time stamp in the trajectory pose sequence, and use this spatial angle as the attitude-trajectory matching degree; If the spatial angle is not greater than the preset spatial angle, that is, the attitude-trajectory matching degree is greater than the preset attitude-trajectory matching degree, then the fusion and collaborative optimization are directly performed based on the trajectory pose sequence. Conversely, the attitude-trajectory matching deviation rate is obtained, and within the null space of the qualified joint configuration corresponding to the current timestamp, the spatial angle is reduced as the optimization objective to perform iterative adjustment, resulting in a new joint configuration and the corresponding end effector attitude. Then, the trajectory pose sequence after iterative adjustment is fused and collaboratively optimized.

9. The redundant robotic arm bagging trajectory generation and attitude planning method as described in claim 8, characterized in that, When the change in rotation angle of any joint in the trajectory pose sequence exceeds the preset change in joint rotation angle, the specific steps of the fusion and collaborative optimization are as follows: The joint configuration and target joint corresponding to the change in rotation angle exceeding the preset change in joint rotation angle are obtained. The time interval corresponding to the change in joint rotation angle exceeding the preset change in joint rotation angle is taken as the abrupt change time interval. The average angular velocity of the joint is calculated based on the target joint rotation angle value at the beginning and end of the interval. If the average angular velocity of a joint exceeds the preset average angular velocity of a joint, the corresponding joint is determined to be in a state of velocity incoordination, and the movement time of the corresponding joint is extended by inserting at least one intermediate time point within the abrupt change time interval. Conversely, the output trajectory pose sequence is used as the final bagging trajectory.

10. The method for generating and planning the trajectory and attitude of a redundant robotic arm for bagging as described in claim 8, characterized in that, When there is no rotation angle change in the trajectory pose sequence that exceeds the preset joint rotation angle change, the specific steps of the fusion and collaborative optimization are as follows: Calculate the absolute value of the deviation between the rotation angle of each joint and the median of the limit in the trajectory pose sequence, and use it as the joint deviation value; If any joint has a joint deviation value that exceeds the corresponding preset joint deviation value, then perform redundancy enhancement posture optimization; otherwise, directly output the trajectory pose sequence as the final bagging trajectory. The redundancy enhancement posture optimization specifically involves: projecting the gradient of the target joint rotation angle moving towards the limit median direction into the null space, optimizing to reduce the joint deviation value, obtaining the optimized trajectory pose sequence, and outputting the optimized trajectory pose sequence as the final bagging trajectory.

Images