Redundant robot motion control method and redundant robot

By decomposing tasks into strict and non-strict tasks, combined with the least squares method, the motion control method of redundant robots solves the problems of complexity and high computational volume of motion control of redundant robots in the prior art, improving accuracy and Reliability while reducing hardware costs.

CN116442210BActive Publication Date: 2024-12-06TENCENT TECHNOLOGY (SHENZHEN) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210021655.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-01-10
Publication Date
2024-12-06
Estimated Expiration
2042-01-10

AI Technical Summary

Technical Problem

The existing motion control methods of redundant robots have problems such as algorithm singularity, large calculation volume, and pseudo-inverse switching of Jacobian matrix when processing multiple tasks, making it difficult to accurately complete tasks and cumbersome procedures.

Method used

A motion control method for redundant robots is proposed. By decomposing tasks into strict tasks and non-strict tasks, high-precision motion parameters are determined separately for strict tasks, and lower-precision motion parameters are determined based on the objective function of non-strict tasks, combined with the minimum The two-multiple method determines the target motion parameters with a lower calculation amount.

Benefits of technology

It improves the accuracy and reliability of motion control, reduces the computing volume and hardware costs, ensures that strict tasks are not affected by non-strict tasks, and enhances the task-taking ability of redundant robots.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116442210B_ABST
    Figure CN116442210B_ABST
Patent Text Reader

Abstract

The present disclosure relates to a motion control method for a redundant robot, which can determine target motion parameters with a relatively low amount of calculation, which can ensure that the strict tasks are accurately executed with high precision on the one hand, and can take the non-strict tasks into consideration on the other hand, under the condition that multiple tasks expected to be implemented by the redundant robot are decomposed into strict tasks and non-strict tasks, and control the redundant robot to move according to the determined target motion parameters, which has the advantages of high motion accuracy and low cost of required computing hardware. In addition, the present invention also relates to a redundant robot motion controller, a redundant robot and a computer program product, which can determine the target motion parameters with a relatively low amount of calculation, and can ensure that the strict tasks are executed with high precision while taking the non-strict tasks into consideration.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the field of robots, and more specifically, to a motion control method for a redundant robot, a redundant robot motion controller capable of executing the motion control method, a redundant robot, and a computer program product. Background Art

[0002] With the development of robot technology, various robots are increasingly used in the fields of smart factories, smart logistics, smart warehousing, etc. to significantly improve the degree of automation and reduce human resource costs. Especially in the field of industrial robots, industrial robots in the form of mechanical arms are usually used to replace manual labor. As a result, in addition to significantly improving the degree of automation of the production line, it also significantly improves production efficiency and processing accuracy.

[0003] Commonly used robotic arms are usually non-redundant robots, that is, robots whose degrees of freedom (number of joints) are less than or equal to the desired task dimension. For example, according to production needs, when the desired task dimension is 6, the traditional six-axis robot is a so-called non-redundant robot. However, if one joint fails, the traditional six-axis robot will no longer be able to reliably complete the task with a task dimension of 6. Therefore, in order to improve the robot's obstacle avoidance ability and reliability, redundant robots (such as seven-axis robots) whose degrees of freedom are greater than or equal to the desired task dimension have been put into use.

[0004] However, the motion control of redundant robots is more complicated than that of non-redundant robots. For multiple tasks to be performed by redundant robots, the existing motion control of redundant robots is mostly based on strict task priority strategies or least squares methods. Here, in the method based on strict task priority strategies, there are problems such as algorithm singularity and large amount of calculation; while in the method based on least squares, tasks of different priorities will affect each other, and the switching of the left pseudo-inverse and right pseudo-inverse of the Jacobian matrix will occur. There are problems such as it is difficult for redundant robots to accurately complete the pre-given tasks and the corresponding program of this method is very cumbersome.

[0005] Therefore, in view of the above-mentioned shortcomings, there is a need for a method that can accurately perform motion control on redundant robots with a low amount of calculation under the condition of considering the priorities of multiple tasks to be implemented by the redundant robots.

[0006] Public Content

[0007] Therefore, the present disclosure provides a motion control method for a redundant robot, which can determine the target motion parameters that can ensure that the strict tasks are accurately executed with high precision and that can take the non-strict tasks into consideration with a low amount of calculation under the condition of considering the priorities of multiple tasks expected to be implemented by the redundant robot, that is, decomposing the multiple tasks expected to be implemented by the redundant robot into strict tasks and non-strict tasks, and control the redundant robot to move according to the determined target motion parameters. Thus, with the above motion control method, on the one hand, multiple tasks are decomposed and operations are performed on the decomposed strict tasks and non-strict tasks respectively, so that the accuracy of the strict tasks is not affected by the non-strict tasks, thereby improving the accuracy and reliability of the motion control method; and on the other hand, when the non-strict tasks contain multiple tasks, the target motion parameters can be determined with a low amount of calculation, thereby reducing the amount of calculation required for the motion control method and thus reducing the hardware cost required to execute the motion control method.

[0008] In addition, the present disclosure also provides a redundant robot motion controller, a redundant robot and a computer program product. It can execute the above motion control method, and therefore has the advantages corresponding to the above motion control method, that is, on the one hand, according to the actual use needs, multiple tasks are decomposed and operations are performed on strict tasks and non-strict tasks respectively, so that the accuracy of strict tasks is not affected by non-strict tasks, thereby improving the accuracy and reliability of the motion control method; and on the other hand, when the non-strict task contains multiple tasks, the target motion parameters can be determined with a lower amount of calculation, thereby reducing the amount of calculation required for the motion control method and thus reducing the hardware cost required to execute the motion control method.

[0009] A first aspect of the present disclosure relates to a motion control method for a redundant robot, wherein the redundant robot includes multiple joints, and the motion control method includes: receiving a task that the redundant robot is expected to achieve; performing task decomposition on the task to decompose the task into a strict task and a non-strict task, wherein the strict task is a task that requires the redundant robot to perform accurately; determining multiple sets of strict task motion parameters, wherein, according to each of the multiple sets of strict task motion parameters, the redundant robot can perform the strict task with a precision higher than a first precision; determining at least one set of non-strict task motion parameters, wherein, according to the non-strict task motion parameters, the redundant robot can perform the non-strict task with a precision higher than a second precision, the first precision being higher than the second precision; and determining a set of target motion parameters based on the multiple sets of strict task motion parameters and the at least one set of non-strict task motion parameters, and controlling the redundant robot to move according to the determined set of target motion parameters.

[0010] Here, the motion control method can consider the priority of multiple tasks to be implemented by the redundant robot, that is, it can decompose multiple tasks into strict tasks that need to be executed with high precision and non-strict tasks that can be executed with lower precision according to the actual needs of the user, and when determining the target motion parameters of the redundant robot required to implement multiple tasks (such as the speed, angular velocity, etc. of each joint of the redundant robot), it is not directly using the least square method for the multiple tasks as a whole to determine the target motion parameters, but determining multiple groups of strict task motion parameters that can execute the strict tasks with the required precision for the strict tasks alone. Thereafter, for example, by "selecting" a group that can take into account non-strict tasks from the multiple groups of strict task motion parameters, a group of target motion parameters can be determined. Thereafter, by controlling each joint of the redundant robot using the determined group of target motion parameters, it is possible to achieve strict assurance of executing strict tasks with high precision while taking into account non-strict tasks. In other words, the motion control method according to the present disclosure can ensure that the precision of strict tasks is not affected by non-strict tasks, thereby improving the accuracy and reliability of the motion control method.

[0011] According to a more detailed implementation of the motion control method of the present disclosure, in the motion control method according to the present disclosure, the strict task motion parameters include at least one of the joint velocity and the joint angular velocity of each joint of the redundant robot that enables the redundant robot to perform the strict task with a precision higher than the first precision, and the non-strict task motion parameters include at least one of the joint velocity and the joint angular velocity of each joint of the redundant robot that enables the redundant robot to perform the non-strict task with a precision higher than the second precision.

[0012] According to a more detailed implementation of the motion control method of the present disclosure, in the motion control method of the present disclosure, the strict task includes a position tracking task of the redundant robot, and the non-strict task includes a posture tracking task of the redundant robot.

[0013] Here, according to the actual production needs of industrial robots, it is usually expected to ensure that the end of the robotic arm of the redundant robot always reaches the spatial position specified in the position tracking task with high precision to accurately process the workpiece. Therefore, the position tracking task is defined as a strict task that needs to be performed with high precision, and the posture tracking task is defined as a non-strict task that allows execution with lower precision. In addition, it is also conceivable that for the use scenario that first needs to ensure that the redundant robot completes the work with an accurate posture, the posture tracking task can also be defined as a strict task that needs to be performed with high precision, and the position tracking task can be defined as a non-strict task that allows execution with lower precision. That is, the specific results of the task decomposition are not limited in this disclosure.

[0014] According to a more detailed implementation of the motion control method disclosed herein, in the motion control method disclosed herein, the non-strict task also includes at least one of a joint angle limit constraint task, an arm angle control task, and a singular avoidance task for each joint of the redundant robot.

[0015] Here, in the motion control method according to the present disclosure, in addition to the posture tracking task, according to the actual production needs of the industrial robot, at least one of the joint angle limit constraint task, arm angle control task and singular avoidance task of each joint expected to be performed by the redundant robot can also be defined as a non-strict task that can be performed with lower precision. In this way, it is possible to take into account the execution of multiple other tasks while ensuring that the strict task is strictly executed with high precision, which significantly improves the task balancing ability of the redundant robot.

[0016] According to a more detailed implementation of the motion control method of the present disclosure, in the motion control method according to the present disclosure, multiple sets of strict task motion parameters are determined, including: determining the multiple sets of strict task motion parameters based on a strict task kinematic model for kinematically describing the strict tasks of the redundant robot.

[0017] Therefore, compared with directly using the least squares method for multiple tasks as a whole, the above separate calculation of strict tasks ensures that the accuracy of strict tasks is not affected by non-strict tasks, thereby improving the accuracy and reliability of the motion control method.

[0018] According to a more detailed implementation of the motion control method of the present disclosure, in the motion control method according to the present disclosure, determining the non-strict task motion parameters includes: determining the non-strict task motion parameters based on an objective function that reflects the difference between the redundant robot motion state and the non-strict task.

[0019] Therefore, compared with directly placing each of the multiple tasks included in the non-strict task in the null space of the strict task, the amount of calculation required to determine the motion parameters of the non-strict task is greatly reduced, and thus the hardware cost required to execute the motion control method is significantly reduced.

[0020] According to a more detailed implementation of the motion control method disclosed in the present invention, in the motion control method disclosed in the present invention, determining the non-strict task motion parameters also includes: based on the objective function, constructing a non-strict task kinematic model for kinematically describing the non-strict task of the redundant robot, and determining the non-strict task motion parameters based on the non-strict task kinematic model.

[0021] As a result, the amount of computation required to determine the non-strict task motion parameters is reduced compared to directly solving the partial differential equations related to the derivatives of the objective function with respect to the non-strict task motion parameters.

[0022] According to a more detailed implementation of the motion control method disclosed in the present invention, in the motion control method disclosed in the present invention, the non-strict task includes multiple non-strict sub-tasks, wherein the determination of the non-strict task motion parameters also includes: in the objective function, setting task weights for the difference between the redundant robot motion state and at least one non-strict sub-task in the non-strict task to distinguish task priorities.

[0023] As a result, a more robust and more reasonable task decomposition of non-strict tasks is achieved, which can better take into account certain non-strict subtasks in the non-strict tasks expected to be implemented by redundant robots according to actual usage requirements.

[0024] However, when constructing a non-strict task kinematic model, i.e., a Jacobian matrix, for kinematically describing a non-strict task, the weighted pseudo-inverse of the non-strict task kinematic model will present different mathematical expressions depending on the number of joints that reach the joint limit positions. In other words, the mathematical expression of the weighted pseudo-inverse of the non-strict task kinematic model is a piecewise function based on the number of joints that reach the joint limit positions.

[0025] In order to calculate the least squares solution of the objective function in a unified solution method, according to a more detailed implementation of the motion control method disclosed herein, in the motion control method disclosed herein, determining the non-strict task motion parameters also includes: performing singular value decomposition on the non-strict task kinematic model to calculate the least squares solution of the objective function, and using the least squares solution as the non-strict task motion parameters.

[0026] In this way, the discontinuity and mutation of joint velocity and joint angular velocity in the solved non-strict task motion parameters are avoided, so that the least squares solution of the non-strict task is not affected by the number of joints reaching the joint limit position.

[0027] According to a more detailed implementation of the motion control method of the present invention, the motion control method of the present invention also includes: after determining a set of target motion parameters, adjusting the strict task kinematic model so that the matrix space corresponding to the non-strict task kinematic model can fall into the orthogonal complementary space corresponding to the strict task kinematic model, and adjusting the target motion parameters based on the adjusted strict task kinematic model.

[0028] Thus, the error between the actual motion state of the redundant robot and the non-strict task when the redundant robot is controlled to perform the above-mentioned multiple tasks is further reduced. In other words, the non-strict task can be performed with higher accuracy through the motion control method according to the present disclosure.

[0029] In summary, according to the motion control method disclosed in the present invention, under the condition of considering the priorities of multiple tasks expected to be implemented by the redundant robot, that is, under the condition of decomposing the multiple tasks expected to be implemented by the redundant robot into strict tasks and non-strict tasks and setting weights for each non-strict task in the non-strict tasks, it is possible to determine the strict task motion parameters that can execute the strict task with the required accuracy for the strict task alone, so that the accuracy of the strict task is not affected by the non-strict task, thereby improving the accuracy and reliability of the motion control method; and on the other hand, after the above-mentioned task decomposition, when the non-strict task contains multiple tasks, for example, by using the least squares method, the motion parameters for executing the non-strict task can be determined with a lower amount of calculation and then placed in the null space of the strict task, rather than directly placing the multiple non-strict subtasks contained in the non-strict task in the null space of the strict task, further reducing the amount of calculation required for the motion control method and thus reducing the hardware cost required to execute the motion control method.

[0030] A second aspect of the present disclosure relates to a redundant robot, the redundant robot comprising: a task decomposition module, the task decomposition module being configured to perform task decomposition on a task so as to decompose the task into a strict task and a non-strict task, wherein the strict task is a task that requires the redundant robot to perform accurately; a strict task motion parameter determination module, the strict task motion parameter determination module being configured to determine a plurality of groups of strict task motion parameters, wherein, according to each of the plurality of groups of strict task motion parameters, the redundant robot can perform the strict task with a precision higher than a first precision; a non-strict task motion parameter determination module, the non-strict task motion parameter determination module being configured to determine a non-strict task motion parameter, wherein, according to the non-strict task motion parameter, the redundant robot can perform the non-strict task with a precision higher than a second precision, the first precision being higher than the second precision; a target motion parameter determination module, the target motion parameter determination module being configured to determine a set of target motion parameters based on the plurality of groups of strict task motion parameters and the at least one group of non-strict task motion parameters; and a joint control module, the joint control module being configured to control the redundant robot to move according to the determined set of target motion parameters.

[0031] Therefore, the redundant robot according to the present disclosure has advantages corresponding to the motion control method according to the present disclosure, that is, the redundant robot according to the present disclosure can ensure that the accuracy of the strict tasks it performs is not affected by non-strict tasks, and has higher accuracy and reliability.

[0032] According to a more detailed embodiment of the redundant robot disclosed in the present invention, in the redundant robot, the strict task motion parameter determination module is also configured to determine the multiple sets of strict task motion parameters based on a strict task kinematic model for kinematically describing the strict task of the redundant robot; and the non-strict task motion parameter determination module is configured to determine the non-strict task motion parameters based on an objective function that reflects the difference between the motion state of the redundant robot and the non-strict task.

[0033] Therefore, compared with the method of directly using the least square method for multiple tasks as a whole, the above separate calculation of the strict task ensures that the accuracy of the strict task is not affected by the non-strict task, thereby improving the accuracy and reliability of the motion control method. In addition, for non-strict tasks, the amount of calculation required to be performed by the redundant robot according to the present disclosure is relatively low, so that the redundant robot according to the present disclosure has a cost advantage in terms of hardware used for operation and storage.

[0034] In summary, the redundant robot according to the present disclosure has advantages corresponding to the motion control method according to the present disclosure, which will not be described in detail for the sake of brevity.

[0035] A third aspect of the present disclosure relates to a redundant robot motion controller, which is configured to execute the motion control method according to the present disclosure according to a received task to control multiple joints in the redundant robot to move. Therefore, the redundant robot motion controller according to the present disclosure has the advantages corresponding to the motion control method according to the present disclosure, which will not be described in detail for the sake of brevity.

[0036] A fourth aspect of the present disclosure relates to a computer program product, the computer program product being stored on a computer-readable storage medium and comprising a computer program or instructions, the computer program or instructions implementing the steps of the motion control method according to the present disclosure when executed by a processor. Thus, the computer program product according to the present disclosure has advantages corresponding to the motion control method according to the present disclosure, which will not be described in detail for the sake of brevity.

[0037] A fifth aspect of the present disclosure relates to a redundant robot, the redundant robot comprising: a plurality of joints, and a redundant robot motion controller, the redundant robot motion controller being configured to execute the motion control method according to the present disclosure according to a received task, so as to control the plurality of joints in the redundant robot to move. Thus, the redundant robot according to the present disclosure has advantages corresponding to the motion control method according to the present disclosure, which will not be described in detail for the sake of brevity. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some exemplary embodiments of the present disclosure, and a person of ordinary skill in the art can obtain other drawings based on these drawings without creative work.

[0039] Herein, in the accompanying drawings:

[0040] Figure 1 A schematic flow chart of a motion control method according to the present disclosure is shown;

[0041] Figure 2 A schematic diagram showing the seven-DOF arm angle is shown;

[0042] Figure 3 A schematic flow chart showing the steps of determining non-rigid task motion parameters;

[0043] Figure 4 A flowchart showing another embodiment of the motion control method according to the present disclosure;

[0044] Figure 5 shows a module diagram of a redundant robot according to the present disclosure;

[0045] Figure 6 A redundant robot according to the present disclosure is shown. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present disclosure more obvious, the exemplary embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments of the present disclosure, and it should be understood that the present disclosure is not limited to the exemplary embodiments described here.

[0047] Furthermore, in the present specification and the drawings, steps and elements having substantially the same or similar features are denoted by the same or similar reference numerals, and repeated descriptions of these steps and elements will be omitted.

[0048] In addition, in this specification and the accompanying drawings, elements are described in singular or plural form, depending on the embodiment. However, the singular and plural forms are appropriately selected for the proposed situation only for the convenience of explanation and are not intended to limit the present disclosure thereto. Therefore, the singular form may include the plural form, and the plural form may also include the singular form, unless the context clearly indicates otherwise.

[0049] In addition, in this specification and the drawings, the terms "first\second" involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It is understandable that "first\second" can be interchanged with a specific order or sequence where permitted, so that the embodiments of the present disclosure described herein can be implemented in an order other than that illustrated or described herein.

[0050] In addition, in this specification and the accompanying drawings, if a flow chart is used to illustrate the steps of the method according to the embodiment of the present disclosure, it should be understood that the previous or subsequent steps are not necessarily performed precisely in order. Instead, various steps can be processed in reverse order or simultaneously, unless explicitly limited by the embodiment of the present disclosure. At the same time, other operations can also be added to these processes, or one or more steps can be removed from these processes.

[0051] As an example, the present disclosure can be applied to the field of defect detection combined with artificial intelligence (AI). Among them, artificial intelligence is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Artificial intelligence is to study the design principles and implementation methods of various intelligent machines so that the machines have the functions of perception, reasoning and decision-making.

[0052] Artificial intelligence technology is a comprehensive discipline that covers a wide range of fields, including both hardware-level and software-level technologies. The basic technologies of artificial intelligence generally include sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics and other technologies. Artificial intelligence software technology mainly includes computer vision technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0053] At present, with the research and progress of artificial intelligence technology, artificial intelligence technology has been studied and applied in many fields, such as common smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, unmanned driving, automatic driving, drones, robots, smart medical care, smart customer service, etc. At present, using the perception, reasoning and decision-making functions of artificial intelligence, artificial intelligence has been combined with robots and applied to smart factories, smart logistics, smart warehousing and other fields to achieve the purpose of replacing manual labor to perform the desired operations on various objects in the above fields and significantly improve the degree of automation in the above fields and reduce human resource costs. Especially in the field of industrial robots, industrial robots in the form of mechanical arms are usually used to replace manual labor. As a result, in addition to significantly improving the degree of automation of the production line, it also significantly improves production efficiency and processing accuracy.

[0054] In addition to the common robotic arms that are usually non-redundant robots, in order to improve the robot's obstacle avoidance ability and reliability, redundant robots whose degrees of freedom are greater than or equal to the desired task dimension have been put into use. However, the motion control of redundant robots is more complicated than that of non-redundant robots. For multiple tasks to be implemented by redundant robots, the existing motion control of redundant robots is mostly based on a strict task priority strategy, that is, the multiple tasks are prioritized and low-priority tasks are placed in the null space of high-priority tasks to enable redundant robots to accurately perform low-priority tasks and high-priority tasks. However, this motion control method has algorithm singularity. And when the task is graded, the null space of multi-level tasks will cause the amount of calculation to increase. Another existing redundant robot motion control method is to use a weighted least squares method, that is, for example, to construct an objective function that reflects the difference between the robot's kinematic state and each task to be implemented by the redundant robot, and use the least squares solution of the objective function as a control parameter. This method has a small amount of calculation, but different priority tasks in this method will affect each other, and according to the limit angle reached by different joint angles, the switching of the left pseudo-inverse and right pseudo-inverse of the Jacobian matrix will occur.

[0055] Therefore, according to a first aspect of the present disclosure, a motion control method for a redundant robot is provided. Figure 1 A schematic flow chart of a motion control method 100 according to the present disclosure is shown. According to an embodiment of the present disclosure, the redundant robot includes a plurality of joints, and the motion control method 100 includes: in step S110, receiving a task that the redundant robot is expected to achieve; in step S120, performing the task on the redundant robot; Perform task decomposition to break down the task into strict tasks With non-strict tasks Among them, the strict task For tasks that require the redundant robot to perform accurately; in step S130, multiple sets of strict task motion parameters are determined According to each of the plurality of sets of strict task motion parameters, the redundant robot can perform the strict task with a precision higher than the first precision. In step S140, at least one set of non-rigid task motion parameters is determined According to the non-strict task motion parameters The redundant robot is capable of performing non-critical tasks with a precision higher than a second precision The first precision is higher than the second precision; and in step S150, based on the multiple sets of strict task motion parameters and the at least one set of non-rigid task motion parameters Determine a set of target motion parameters In step S160, according to the determined set of target motion parameters The redundant robot is controlled to move.

[0056] Here, the motion control method 100 can consider the priorities of multiple tasks to be implemented by the redundant robots, that is, it can prioritize the tasks according to the actual needs of the user. Break it down into strict tasks that need to be performed with high precision and non-strict tasks that allow execution with lower precision And in determining the target motion parameters of redundant robots required to achieve multiple tasks When calculating the speed and angular velocity of each joint of a redundant robot, it is not necessary to directly use the least square method for multiple tasks as a whole to determine the target motion parameters. But only for strict tasks Determine that the rigorous task can be performed with the required accuracy Multiple sets of strict task motion parameters Thereafter, for example, by selecting from the plurality of sets of strict task motion parameters "Select" those who can take care of non-strict tasks A set of target motion parameters can be determined Thereafter, using the determined set of target motion parameters Controlling the joints of redundant robots ensures that strict tasks are performed with high precision. While being able to take care of non-rigorous tasks In other words, the motion control method 100 according to the present disclosure can ensure strict task The accuracy is not affected by non-rigorous tasks The accuracy and reliability of the motion control method 100 are improved.

[0057] According to a more detailed embodiment of the motion control method 100 of the present disclosure, in the motion control method 100 of the present disclosure, the strict task motion parameter comprising enabling the redundant robot to perform the critical task with a precision greater than a first precision At least one of the joint velocity and joint angular velocity of each joint of the redundant robot, and the non-strict task motion parameters The method includes enabling the redundant robot to perform the non-critical task with a precision higher than a second precision. At least one of the joint velocity and the joint angular velocity of each joint of the redundant robot.

[0058] Here, the strict task motion parameters Or the non-strict task motion parameters That is, to enable redundant robots to perform tasks with strict or non-strict tasks Completing rigorous tasks with corresponding precision or non-strict tasks At least one of the joint velocity and the joint angular velocity. Based on the multiple sets of strict task motion parameters and the at least one set of non-rigid task motion parameters A set of target motion parameters determined Each of the joint speeds and joint angular velocities of each joint of the redundant robot is also at least one of the joint speeds and joint angular velocities of each joint of the redundant robot. Thus, after determining the joint speeds and joint angular velocities of each joint of the redundant robot, i.e., a set of target motion parameters Then, according to the target motion parameters By driving the redundant robot to move, the desired strict task of the redundant robot can be completed. and non-strict tasks

[0059] According to a more detailed implementation of the motion control method 100 of the present disclosure, in the motion control method 100 of the present disclosure, the strict task Including the position tracking task of the redundant robot And the non-strict task Including the posture tracking task of the redundant robot

[0060] The following is an exemplary illustration of the above position tracking task: And posture tracking task kinematic representation of .

[0061] In the motion control method 100 according to the present disclosure, the Cartesian kinematics of the redundant robot can be expressed as follows:

[0062]

[0063] Where x∈R m is the redundant robot Cartesian position and posture, is the Cartesian speed of the redundant robot, i.e. the task that the redundant robot is expected to achieve, received in step S110, are the joint velocities and joint angular velocities of the redundant robot, J is the Jacobian matrix of the redundant robot for kinematic description, and m is the task dimension to be achieved, and n is the degree of freedom (ie, the number of joints) of the redundant robot.

[0064] In step S120, the task For example, tasks can be decomposed into strict tasks. Position tracking task and as a non-strict task The posture tracking task Its matrix form is as follows:

[0065] x=[x1,x2] T (2)

[0066] Therefore, the position tracking task And posture tracking task The kinematic representation of is as follows:

[0067]

[0068]

[0069] Among them, J1 is the Jacobian matrix for kinematic description of position tracking task, and J2 is the Jacobian matrix for kinematic description of attitude tracking task. is the motion parameter of the position tracking task, Motion parameters for the pose tracking task.

[0070] Here, according to the actual production needs of industrial robots, it is usually expected to ensure that the end of the redundant robot's arm always reaches the spatial position specified in the position tracking task with high precision to accurately process the workpiece; and if some joints of the redundant robot reach the limit position, it is expected to ensure the accurate execution of the position tracking task under the condition of losing the execution accuracy of the posture tracking task to continue to accurately process the workpiece. Defined as a rigorous task that needs to be performed with high precision And the posture tracking task Defined as non-strict tasks that can be performed with lower precision In addition, it is also conceivable that the posture tracking task can also be used for use scenarios where it is necessary to first ensure that the redundant robot completes the work in the correct posture Defined as a rigorous task that needs to be performed with high precision And the location tracking task Defined as non-strict tasks that can be performed with lower precision That is, the present disclosure does not limit the specific results of task decomposition.

[0071] According to a more detailed implementation of the motion control method 100 of the present disclosure, in the motion control method 100 according to the present disclosure, the non-strict task also includes at least one of a joint angle limit constraint task b, an arm angle control task and a singular avoidance task for each joint of the redundant robot.

[0072] The following is an exemplary kinematic representation of the joint angle limit constraint task b for each joint of the redundant robot.

[0073] In the joint angle limit constraint task, the joint angle limit constraint task is set for each joint separately. The joint angle limit constraint task b for the i-th joint i The kinematics is expressed as follows:

[0074]

[0075] Where r is the total number of kinematic constraints and i = 1, 2, 3, ..., r-1, ρ i is the kinematic function, which can be expressed as:

[0076]

[0077] in, and are the upper and lower joint limits of the i-th joint, and Δ i is i The boundary value of .

[0078] Here, as shown in the above formula (5), the joint angle limit constraint task for the i-th joint is configured as follows: when the joint corresponding to the joint angle limit constraint task does not reach the limit, that is, ρ i ≥Δ i , then the motion state of the joint at the next moment t+1 is Maintain the motion state F at the previous moment t i | t , that is, b i |t+1 =F i | t ; and when the joint corresponding to the joint angle limit constraint task reaches the limit, that is, ρ i <Δ i , then the motion state b of the joint at the next moment t+1 i | t+1 is 0, i.e., the movement of the joint is stopped.

[0079] The following combination Figure 2 , which exemplarily shows the kinematic representation of the arm angle control task.

[0080] Figure 2 Schematically shows a schematic diagram of the seven-degree-of-freedom arm angle. Figure 2 As shown, the angle between plane O1O4O7 and a reference plane is generally regarded as the arm angle. Figure 2 In the equation, p is a vector perpendicular to w (i.e., the vector between joints 1 and 7) in the plane O1O4O7, and l is a vector perpendicular to w in the reference plane. V is a coordinate axis in the reference plane. Here, the angle between p and l is usually referred to as the arm angle. Therefore, the arm angle can be expressed as follows:

[0081] ψ=ψ(q)=Atan2(w T (V×p),V T p) (7)

[0083] Therefore, the differential expression of the arm angle of the redundant robot, that is, the arm angle control task As shown below:

[0084]

[0085] Among them, J ψ (q) is the arm angle dependent Jacobian matrix.

[0086] The arm angle can directly reflect the configuration of the redundant robot arm. In addition, it is also conceivable that the actual selected arm angle can also be expanded to any other parameter reflecting the configuration of the robot arm. For example, if a certain joint can remain unchanged during the overall movement, the angle of the joint to one can be selected as the generalized "arm angle".

[0087] In summary, in the motion control method 100 according to the present disclosure, in addition to the posture tracking task In addition, according to the actual production needs of industrial robots, the joint angle limit constraint tasks b and arm angle control tasks of each joint expected to be performed by redundant robots can also be set. and singular avoidance tasks are defined as non-strict tasks that can be performed with lower accuracy And thus determine the ability to strictly ensure that strict tasks are performed with high precision It can also take into account the joint angle limit constraint task b and the arm angle control task Thus, in the motion control method 100 according to the present disclosure, it is possible to take into account the execution of multiple other tasks while ensuring that strict tasks are strictly executed with high precision, thereby significantly improving the task-compatibility capability of the redundant robot.

[0088] According to a more detailed embodiment of the motion control method 100 of the present disclosure, in the motion control method 100 of the present disclosure, when determining multiple sets of strict task motion parameters The step S130 includes: determining the plurality of sets of strict task motion parameters based on a strict task kinematics model for kinematically describing the strict task of the redundant robot.

[0089] The following example shows the position tracking task: As a strict task In this case, the position tracking task parameters Strict task motion parameters

[0090] As shown in the above formula (3), the position tracking task The kinematic representation of is as follows:

[0091]

[0092] Therefore, in determining multiple sets of strict task motion parameters In step S130, the inverse kinematics calculation can be used to determine the position tracking task. The joint speed and joint angular velocity of the redundant robot, that is, the position tracking task parameters

[0093]

[0094] in, is the null space of the redundant robot. ε1 is an arbitrary vector in the null space. In other words, through the above formula (10), multiple groups that can achieve the above position tracking task are determined Position tracking task parameters

[0095] Here, for the strict tasks obtained through task decomposition, (For example, position tracking tasks ), the Jacobian matrix (for example, the Jacobian matrix J1 for kinematic description of the position tracking task) can be used to describe the kinematics of the strict task of the redundant robot. Based on the strict task and the Jacobian matrix, with the help of the above robot inverse kinematics calculation, multiple sets of strict task motion parameters (for example, position tracking task parameters) that can perform the strict task with the required accuracy can be determined for the strict task alone. ), that is, the joint speed and joint angular velocity of each joint of the redundant robot. Therefore, compared with the method of directly using the least square method for multiple tasks as a whole, the above separate calculation of the strict task ensures that the accuracy of the strict task is not affected by the non-strict task, thereby improving the accuracy and reliability of the motion control method 100.

[0096] For example, in a non-strict task Only pose tracking tasks are included In the case of, as shown in the above formula (4), the posture tracking task The kinematic representation of is as follows:

[0097]

[0098] Therefore, through inverse kinematics calculation, the The joint speed and joint angular velocity of the redundant robot, that is, the posture tracking task parameters

[0099]

[0100] in, is the null space of the redundant robot. ε2 is an arbitrary vector in the null space. In other words, through the above formula (12), multiple groups that can achieve the above posture tracking task are determined Parameters of the posture tracking task

[0101] Here, by placing non-strict tasks Included posture tracking tasks Placed on strict tasks For example, position tracking tasks In the null space of :

[0102]

[0103] The vector in the null space can be found:

[0104]

[0105] Substituting the above vector ε1 back into formula (10), we can obtain and posture tracking tasks Target motion parameters

[0106]

[0107] However, as mentioned above, for non-rigid tasks Not only posture tracking tasks It also includes, for example, joint angle limit constraint tasks b, arm angle control tasks In the case of multiple other tasks, the above method of directly placing each of the multiple tasks included in the non-strict task in the null space of the strict task will disadvantageously lead to a significant increase in the amount of calculation.

[0108] To solve the above technical problems, according to a more detailed implementation of the motion control method 100 disclosed in the present invention, in the motion control method 100 disclosed in the present invention, when determining the non-strict task motion parameters The step S140 includes: based on the motion state of the redundant robot and the non-strict task The objective function L of the difference determines the non-rigid task motion parameters

[0109] Here, for example, in the non-strict task Also includes posture tracking tasks Joint angle limit constraint task b for each joint of the redundant robot, arm angle control task and singular avoidance tasks, reflecting the redundant robot motion state and non-strict tasks The objective function L of the difference can be constructed as:

[0110]

[0111] Among them, the first term in the objective function L Reflects the redundant robot motion state and posture tracking task The difference, the second It reflects the difference between the motion state of the redundant robot and the joint angle limit constraint task b of each joint. The third item Reflects the redundant robot motion state and arm angle control task The difference, and the fourth term Mirrors the oddball avoidance task.

[0112] Here, based on the objective function L, for example, by solving the optimal solution of the objective function, that is, for example, by solving the following partial differential equation:

[0113]

[0114] The non-strict task motion parameters that can take into account the multiple other tasks included in the non-strict task can be solved. Similarly, by placing it in the strict task For example, position tracking tasks In the null space of :

[0115]

[0116] The corresponding vector ε1 can be solved. Similarly, by substituting the vector ε1 back into formula (10), a set of target motion parameters can be determined that can strictly ensure the execution of strict tasks with high precision while taking into account multiple other tasks such as joint angle limit constraint tasks and arm angle control tasks. Thus, compared to directly placing each of the multiple tasks included in the non-strict task in the null space of the strict task, the amount of calculation required to determine the motion parameters of the non-strict task is greatly reduced, and thus the hardware cost required to execute the motion control method 100 is significantly reduced. In addition, it should be emphasized that the representation of the objective function L according to formula (16) is only an exemplary representation, which should not be understood as limiting. If the non-strict task also includes other tasks, then based on the concept of the present disclosure, the above objective function L can also include corresponding items reflecting the differences with the other tasks.

[0117] In order to find the optimal solution of the objective function, in addition to tediously solving the partial differential equation shown in formula (17), other methods can also be used. Figure 3 As shown, according to a more detailed implementation of the motion control method 100 of the present disclosure, in the motion control method 100 of the present disclosure, when determining the non-strict task motion parameters In step S140, the method further includes: in sub-step S142, based on the objective function L, constructing a non-strict task for the redundant robot Non-rigid task kinematics model for kinematic description C ; and in sub-step S143, based on the non-strict task kinematics model J C , determine the non-strict task motion parameters.

[0118] In addition, if Figure 3 As shown in the dashed box, according to a more detailed implementation of the motion control method 100 of the present disclosure, in the motion control method 100 of the present disclosure, the non-strict task includes a plurality of non-strict subtasks, wherein in determining the motion parameters of the non-strict task In step S140, it also includes: in the sub-step S141 shown in the dotted box, in the objective function L, for the redundant robot motion state and the non-strict task The task weight w is set based on the difference of at least one non-strict subtask in the task to distinguish the task priorities.

[0119] Here, for example, a task weight w is set for the joint angle limit constraint task and the arm angle control task for each joint. After setting the task weight w, the objective function L can be constructed as follows:

[0120]

[0121] Among them, the rth kinematic constraint is regarded as the arm angle, so w i is the task weight set for the i-th joint, and w r is the task weight set for the arm angle control task. Here, as shown in formula (19), the larger the task weight set for a joint, the more significant the impact of the difference with the joint angle limit constraint task for the joint on the objective function L. In other words, when solving the non-strict task motion parameters by the least squares method, The non-rigid task motion parameters determined It will be more inclined to "take into account" the joint angle limit constraint task for the joint. As a result, a more robust and more reasonable task decomposition of non-strict tasks is achieved, and some non-strict subtasks in the non-strict tasks expected to be achieved by the redundant robot can be better taken into account according to actual usage requirements.

[0122] Here, the task weight w can be expressed in the form of a weight matrix:

[0123] w=diag(w1,w2,…,w r ) (20)

[0124]

[0125] Here, the weighted non-strict task It can be expressed as:

[0126]

[0127] And, the unweighted non-rigid task kinematic model J C It can be expressed as:

[0128]

[0129] And, the weighted non-rigid task kinematic model J wC It can be expressed as:

[0130]

[0131] Thus, according to equations (22), (23) and (24), the kinematics of the non-strict task corresponding to the objective function L shown in equation (19) can be expressed as follows:

[0132]

[0133] Therefore, by using the inverse kinematics operation, the non-rigid task motion parameters can be determined

[0134]

[0135] Here, It's J wC The weighted pseudo-inverse of . And its specific expression is as follows:

[0136]

[0137] Wherein, m is the task dimension that is expected to be achieved, n is the degree of freedom of the redundant robot (i.e., the number of joints), and k is the number of joints in the redundant robot that reach the joint limit.

[0138] Similarly, by placing non-strict tasks within strict tasks For example, position tracking tasks In the null space of :

[0139]

[0140] The corresponding vector ε1 can be solved. Similarly, by substituting the vector ε1 back into formula (10), a set of target motion parameters can be determined that can strictly ensure the execution of strict tasks with high precision while taking into account multiple other tasks such as joint angle limit constraint tasks and arm angle control tasks. According to a set of target motion parameters determined By controlling the redundant robot to move, the redundant robot can ensure that the rigorous tasks are performed with high precision while taking into account non-rigid tasks.

[0141] In summary, in order to find the optimal solution of the objective function, in the motion control method 100 according to the present disclosure, a non-strict task kinematic model for kinematic description of the non-strict task can be constructed based on the objective function. The non-strict task kinematic model can also be constructed with the help of the Jacobian matrix J wC Then, as described above, with the help of the robot inverse kinematics calculation, it is possible to determine the robot that can simultaneously take into account non-strict tasks. Non-strict task motion parameters for multiple tasks in Therefore, compared with the direct solution of the objective function L with respect to the non-strict task motion parameters The partial differential equations related to the derivatives of , reduce the determination of non-rigid task motion parameters The amount of computation required.

[0142] However, when constructing a non-strict task kinematic model, i.e., the Jacobian matrix, for describing the kinematics of a non-strict task, as shown in formula (27), the weighted pseudo-inverse of the non-strict task kinematic model is will present a different mathematical expression. In other words, the weighted pseudo-inverse of the non-rigid task kinematic model The mathematical expression of is a piecewise function based on the number k of joints that reach the joint limit position.

[0143] In order to calculate the least squares solution of the objective function in a unified solution method, Figure 3 As shown in the dashed box, according to a more detailed implementation of the motion control method 100 of the present disclosure, in the motion control method 100 of the present disclosure, in the determination of the non-strict task motion parameters The step S140 further includes: in a sub-step S142' shown in a dotted box, the non-strict task kinematic model J wC Perform singular value decomposition to calculate the least squares solution of the objective function L, and use the least squares solution as the non-rigid task motion parameter

[0144] Here, J wC Performing singular value decomposition, we can get:

[0145]

[0146] From this, we can determine its weighted pseudo-inverse The unified representation of is:

[0147]

[0148] Therefore, by substituting the above formula (30) into formula (26), the non-strict task motion parameters can be determined

[0149]

[0150] Similarly, by placing non-strict tasks within strict tasks For example, position tracking tasks In the null space of :

[0151]

[0152] The corresponding vector ε1 can be solved:

[0153]

[0154] Similarly, by substituting the vector ε1 back into formula (10), a set of target motion parameters can be determined that can strictly ensure the execution of strict tasks with high precision while taking into account multiple other tasks such as joint angle limit constraint tasks and arm angle control tasks.

[0155]

[0156] In summary, by performing singular value decomposition on the non-strict task kinematic model, a unified solution method is used to calculate the least square solution of the objective function L, i.e., the non-strict task motion parameters In this way, the discontinuity and mutation of joint velocity and joint angular velocity in the solved non-strict task motion parameters are avoided, so that the least squares solution of the non-strict task is not affected by the number k of joints reaching the joint limit position.

[0157] However, when a set of target motion parameters determined as in formula (34) Substituting back into formula (25), we can find that it is similar to the weighted non-strict task There is the following error R E2 :

[0158]

[0159] In other words, adopting the motion control method 100 according to the present disclosure will result in a weighted non-rigid task The residual is not 0.

[0160] In order to eliminate the above error R E2 ,like Figure 4 As shown, according to a more detailed implementation of the motion control method 100 of the present disclosure, the motion control method 100 further includes: after the step S150 of determining a set of target motion parameters, in step S170, adjusting the strict task kinematics model J1 so that the matrix space corresponding to the non-strict task kinematics model can fall into the orthogonal complement space corresponding to the strict task kinematics model, and adjusting the target motion parameters based on the adjusted strict task kinematics model

[0161] At this time, the error R of the non-strict task E2 for:

[0162]

[0163] Here, weighted processing is used The strict task kinematics model J1 is adjusted so that the matrix space corresponding to the non-strict task kinematics model can fall into the orthogonal complementary space corresponding to the strict task kinematics model. After the processing, the motion control law of the redundant robot, i.e., the target motion parameter for:

[0164]

[0165] At this time, in order to make R E2 is 0, we can design the weight W so that To eliminate the error R shown in the above formula (36) E2 .

[0166] Here, assuming because It can be deduced that Therefore, here,

[0167] Therefore, a set of target motion parameters can be obtained that are finally adjusted to perform strict tasks with high precision while taking into account non-strict tasks.

[0168]

[0169] Here, in the motion control method 100 according to the present disclosure, by placing the matrix space corresponding to the non-strict task kinematic model into the orthogonal complementary space corresponding to the strict task kinematic model, and adjusting the target motion parameters accordingly, the error between the actual motion state of the redundant robot and the non-strict task when controlling the redundant robot to perform the above-mentioned multiple tasks is further reduced. In other words, by the motion control method 100 according to the present disclosure, the non-strict task can be executed with higher accuracy.

[0170] In summary, according to the motion control method 100 disclosed in the present invention, under the condition of considering the priorities of multiple tasks expected to be implemented by the redundant robot, that is, under the condition of decomposing the multiple tasks expected to be implemented by the redundant robot into strict tasks and non-strict tasks and setting weights for each non-strict task in the non-strict tasks, it is possible to determine the strict task motion parameters that can execute the strict task with the required accuracy for the strict task alone, so that the accuracy of the strict task is not affected by the non-strict task, thereby improving the accuracy and reliability of the motion control method 100; and on the other hand, after the above-mentioned task decomposition, when the non-strict task contains multiple tasks, for example, by using the least squares method, the motion parameters for executing the non-strict task can be determined with a lower amount of calculation and then placed in the null space of the strict task, rather than directly placing the multiple non-strict subtasks contained in the non-strict task in the null space of the strict task, further reducing the amount of calculation required for the motion control method 100 and thus reducing the hardware cost required to execute the motion control method 100.

[0171] Furthermore, according to a second aspect of the present disclosure, a redundant robot 200 is provided. Figure 5 Schematically shows a module diagram of a redundant robot 200 according to the present disclosure. Here, the redundant robot 200 includes: a task decomposition module 210, the task decomposition module 210 is configured to decompose the task to decompose the task into strict tasks With non-strict tasks Among them, the strict task A strict task motion parameter determination module 220, wherein the strict task motion parameter determination module 220 is configured to determine multiple sets of strict task motion parameters. According to each of the plurality of sets of strict task motion parameters, the redundant robot 220 can perform the strict task with a precision higher than the first precision. The non-strict task motion parameter determination module 230 is configured to determine the non-strict task motion parameter According to the non-strict task motion parameters The redundant robot is capable of performing non-critical tasks with a precision higher than a second precision The first precision is higher than the second precision; a target motion parameter determination module 240, the target motion parameter determination module 240 is configured to determine the target motion parameter based on the multiple sets of strict task motion parameters and the at least one set of non-rigid task motion parameters Determine a set of target motion parameters and a joint control module 250, wherein the joint control module 250 is configured to perform a set of target motion parameters according to the determined The redundant robot 200 is controlled to move.

[0172] Therefore, the redundant robot according to the present disclosure has the advantages corresponding to the motion control method 100 according to the present disclosure, that is, it is possible to consider the priorities of multiple tasks to be implemented by the redundant robot, that is, it is possible to decompose multiple tasks into strict tasks that need to be performed with high precision according to the actual needs of the user. and non-strict tasks that allow execution with lower precision And only for strict tasks Determine multiple sets of critical task motion parameters that can perform the critical task with the required accuracy Thereafter, for example, by selecting from the plurality of sets of strict task motion parameters A set of target motion parameters can be determined by "selecting" a group that can take into account non-strict tasks Therefore, the redundant robot 200 according to the present disclosure can ensure that the accuracy of the strict tasks it performs will not be affected by non-strict tasks, and has higher accuracy and reliability.

[0173] According to a more detailed embodiment of the redundant robot 200 of the present disclosure, in the redundant robot 200, the strict task motion parameter determination module 220 is further configured to determine the strict task motion parameter of the redundant robot 200 based on the strict task motion parameter of the redundant robot 200. A strict task kinematic model for kinematic description is used to determine the multiple sets of strict task motion parameters. The non-strict task motion parameter determination module 230 is configured to reflect the motion state of the redundant robot and the non-strict task The objective function of the difference determines the non-rigid task motion parameters

[0174] Therefore, for strict tasks The redundant robot 200 according to the present disclosure can determine multiple sets of strict task motion parameters capable of executing the strict tasks with the required accuracy for the strict tasks by means of robot inverse kinematics calculation. Therefore, compared with the method of directly using the least square method for multiple tasks as a whole, the above separate calculation of the strict task ensures that the accuracy of the strict task is not affected by the non-strict task, thereby improving the accuracy and reliability of the redundant robot motion. The redundant robot 200 according to the present disclosure uses the least squares method to determine the number of tasks that can be taken into account at the same time. Non-strict task motion parameters for multiple tasks in Therefore, compared with directly placing each of the multiple tasks included in the non-strict task in the null space of the strict task, the redundant robot according to the present invention needs to perform a lower amount of calculations, so that the redundant robot according to the present invention has a cost advantage in terms of hardware used for operation and storage.

[0175] In summary, the redundant robot according to the present disclosure has the advantages corresponding to the motion control method 100 according to the present disclosure, which will not be described in detail for the sake of brevity.

[0176] In addition, according to a third aspect of the present disclosure, a redundant robot motion controller is provided, which is configured to execute the motion control method 100 according to the present disclosure according to the received task to control multiple joints in the redundant robot to move. Therefore, the redundant robot motion controller according to the present disclosure has the advantages corresponding to the motion control method 100 according to the present disclosure, which will not be described in detail for the sake of brevity.

[0177] In addition, according to a fourth aspect of the present disclosure, a computer program product is provided, the computer program product being stored on a computer-readable storage medium and comprising a computer program or instructions, the computer program or instructions implementing the steps of the motion control method 100 according to the present disclosure when executed by a processor. For more details on the functions of the computer program product, reference may be made to the above description of the motion control method 100 described in the first aspect of the present disclosure. Thus, the computer program product according to the present disclosure has the advantages corresponding to the motion control method 100 according to the present disclosure, which will not be described in detail for the sake of brevity.

[0178] Furthermore, according to a fifth aspect of the present disclosure, a redundant robot 300 is provided. Figure 6 A redundant robot 300 according to the present disclosure is shown. The redundant robot 300 includes: a plurality of joints 310, and a redundant robot motion controller 320, wherein the redundant robot motion controller is configured to execute the motion control method 100 according to the present disclosure according to the received task, so as to control the movement of the plurality of joints 300 in the redundant robot. The redundant robot motion controller 320 can be implemented as any device that can execute the motion control method as described in the first aspect of the present disclosure, including but not limited to FPGA, DSP, ARM single-chip microcomputer, CPU, etc. For more details on the functions of the redundant robot motion controller 310, reference can be made to the explanation of the motion control method 100 described in the first aspect of the present disclosure above. Therefore, the redundant robot 300 according to the present disclosure has the advantages corresponding to the motion control method 100 according to the present disclosure, which will not be described in detail here for the sake of brevity.

[0179] The exemplary embodiments of the present disclosure described in detail above are merely illustrative and not restrictive. It should be understood by those skilled in the art that various modifications and combinations may be made to these embodiments or their features without departing from the principles and spirit of the present disclosure, and such modifications should fall within the scope of the present disclosure.

Claims

1. A motion control method for a redundant robot, wherein: The redundant robot includes a plurality of joints, and the motion control method includes: receiving a task that the redundant robot is expected to perform; Decomposing the task into strict tasks and non-strict tasks, wherein the strict tasks are tasks that require the redundant robot to perform accurately; determining a plurality of sets of strict task motion parameters, wherein, according to each of the plurality of sets of strict task motion parameters, the redundant robot is capable of performing the strict task with an accuracy higher than a first accuracy; determining at least one set of non-strict task motion parameters, wherein according to the non-strict task motion parameters, the redundant robot can perform the non-strict task with a precision higher than a second precision, the first precision being higher than the second precision; Based on the multiple groups of strict task motion parameters and the at least one group of non-strict task motion parameters, a group of target motion parameters is determined, and the redundant robot is controlled to move according to the determined group of target motion parameters.

2. The motion control method according to claim 1, wherein: The strict task motion parameter includes at least one of a joint velocity and a joint angular velocity of each joint of the redundant robot that enables the redundant robot to perform the strict task with an accuracy higher than a first accuracy, and The non-strict task motion parameter includes at least one of a joint velocity and a joint angular velocity of each joint of the redundant robot that enables the redundant robot to perform the non-strict task with a precision higher than a second precision.

3. The motion control method according to claim 1, wherein: The strict task includes a position tracking task for the redundant robot, and The non-strict task includes a posture tracking task of the redundant robot.

4. The motion control method according to claim 3, wherein: The non-strict task also includes at least one of a joint angle limit constraint task, an arm angle control task, and a singular avoidance task for each joint of the redundant robot.

5. The motion control method according to any one of claims 1 to 4, wherein: Determine multiple sets of strict mission motion parameters, including: The plurality of groups of strict task motion parameters are determined based on a strict task kinematics model for kinematically describing the strict tasks of the redundant robot.

6. The motion control method according to claim 5, wherein: Determine non-mission-critical motion parameters, including: The non-strict task motion parameters are determined based on an objective function reflecting the difference between the redundant robot motion state and the non-strict task.

7. The motion control method according to claim 6, wherein: Determine non-critical task motion parameters, including: Based on the objective function, a non-strict task kinematic model is constructed to describe the non-strict tasks of the redundant robot kinematically, and Based on the non-strict task kinematic model, non-strict task motion parameters are determined.

8. The motion control method according to claim 7, wherein: The non-strict task includes a plurality of non-strict subtasks. Wherein, the determining of non-strict task motion parameters further includes: In the objective function, a task weight is set according to the difference between the motion state of the redundant robot and at least one non-strict subtask in the non-strict task to distinguish the task priority.

9. The motion control method according to claim 7, wherein: Determine non-critical task motion parameters, including: Singular value decomposition is performed on the non-strict task kinematic model to calculate a least square solution of the objective function, and the least square solution is used as the non-strict task kinematic parameters.

10. The motion control method according to claim 7, further comprising: After determining a set of target motion parameters, the strict task kinematic model is adjusted so that the matrix space corresponding to the non-strict task kinematic model can fall into the orthogonal complementary space corresponding to the strict task kinematic model, and Based on the adjusted strict task kinematic model, the target motion parameters are adjusted.

11. A redundant robot, comprising: A task decomposition module, wherein the task decomposition module is configured to perform task decomposition on a task so as to decompose the task into a strict task and a non-strict task, wherein the strict task is a task that requires the redundant robot to perform accurately; a strict task motion parameter determination module, the strict task motion parameter determination module being configured to determine a plurality of sets of strict task motion parameters, wherein, according to each of the plurality of sets of strict task motion parameters, the redundant robot is capable of performing the strict task with an accuracy higher than a first accuracy; a non-strict task motion parameter determination module, the non-strict task motion parameter determination module being configured to determine at least one set of non-strict task motion parameters, wherein according to the non-strict task motion parameters, the redundant robot is capable of performing the non-strict task with a precision higher than a second precision, the first precision being higher than the second precision; a target motion parameter determination module configured to determine a set of target motion parameters based on the plurality of sets of strict task motion parameters and the at least one set of non-strict task motion parameters; and A joint control module is configured to control the redundant robot to move according to a determined set of target motion parameters.

12. The redundant robot according to claim 11, wherein: The strict task motion parameter determination module is further configured to determine the plurality of sets of strict task motion parameters based on a strict task kinematics model for kinematically describing the strict tasks of the redundant robot; and The non-strict task motion parameter determination module is configured to determine the non-strict task motion parameter based on an objective function reflecting the difference between the redundant robot motion state and the non-strict task.

13. A redundant robot motion controller, configured to execute the motion control method according to any one of claims 1 to 10 according to a received task, so as to control the motion of multiple joints in the redundant robot.

14. A computer program product, stored on a computer-readable storage medium and comprising a computer program or instructions, which, when executed by a processor, implement the steps of the motion control method according to any one of claims 1 to 10.

15. A redundant robot, comprising: Multiple joints, and A redundant robot motion controller, wherein the redundant robot motion controller is configured to execute the motion control method according to any one of claims 1 to 10 according to a received task, so as to control the motion of multiple joints in the redundant robot.