A picking robot sliding preset performance control method, storage medium and control system

Abstract

The present invention relates to a sliding preset performance control method, storage medium and control system for a picking robot, including real-time acquisition of a target position during the robot picking process; acquisition of position information, speed information and torque information of the robot joints; designing a control law to offset errors in gravity, Coriolis force and centrifugal force; introducing sliding preset performance control, and combining the control law to compensate for errors in a robot model calculated using estimated values ​​of inertial parameters; and optimizing the torque input of the joints during the robot picking process based on the error compensation results. The present invention provides a picking path consisting of multiple position points, with each position point on the picking path as a desired position, and continuously controlling the robot end effector to move along the position points on the picking path, thereby controlling the robot to move along the picking path. By controlling the speed of the joints, the robot end effector moves at a desired speed, thereby ensuring smooth operation of the end and reducing the occurrence of jitter.

Description

Technical Field

[0001] The present invention relates to the field of robot control technology, and in particular to a method for controlling sliding preset performance of a picking robot, a storage medium and a control system. Background Art

[0002] Harvesting robots are increasingly being used in smart agriculture, thanks to their superior precision, adaptive control capabilities, visual target recognition systems, and non-destructive harvesting. Fruit and vegetable harvesting robots are a prime example. Fruit and vegetable harvesting operations (including the non-destructive harvesting of apples and grapes) require precise positioning of the target and flexible cutting of the fruit stems. This requires millimeter-level control of the robot's end effector to ensure no damage to the target skin and a smooth cut on the stem. This places extremely high demands on the intelligent execution system. Harvesting robots, such as the "Smart Orchard," leverage precise positioning, multi-sensor fusion environmental perception, and flexible force control technology to significantly improve harvesting success rates and efficiency in complex foliage environments.

[0003] However, the inventors discovered that conventional robots experience significant tracking error jitter before reaching steady state during the harvesting process. Furthermore, once parameters are selected, the PCB (performance constraint boundary) is fixed. Changes to the initial state (or reference signal) require re-verification to ensure that the initial error still meets the initial constraint. This can cause the robot to lose its target during the harvesting process. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method, storage medium and control system for controlling the sliding preset performance of a picking robot to overcome the deficiencies in the above-mentioned prior art.

[0005] The present invention solves the above-mentioned technical problem with the following technical solution: A method for controlling the sliding preset performance of a picking robot, comprising the following steps:

[0006] Step S01: obtaining the target position of the robot during the picking process in real time;

[0007] Step S02: Acquire the position information, speed information and torque information of the robot joints;

[0008] Step S03: Designing a control law to offset the errors of gravity, Coriolis force and centrifugal force;

[0009] Step S04: introducing sliding preset performance control to the robot model, combining the control law, and compensating for the error of the robot model calculated using the estimated inertia parameters;

[0010] Step S05: Optimize the torque input of the joints during the robot picking process based on the error and the error compensation result.

[0011] On the basis of the above technical solution, the present invention can also be improved as follows.

[0012] Furthermore, step S01 specifically includes:

[0013] The target position is captured by the picking target determination module, and the expected parameters of each joint of the robot are calculated through the robot inverse kinematics model. Each joint moves according to the expected parameters to form an expected trajectory, so that the end of the robot reaches the expected posture.

[0014] Furthermore, the expression of the control law designed in step S03 is:

[0015] ;

[0016] Where, is the robot joint angle, is the robot joint angular velocity, is the inertia matrix, and the joint angle Related, is the Coriolis force and centrifugal force matrix, and the joint angle and robot joint angular velocity Related, is the gravity vector, and the joint angle Related, is the joint torque, Slide preset performance control input for the design.

[0017] Furthermore, the expression of the robot model in step S04 is:

[0018] ;

[0019] Where, is the robot joint angle, is the robot joint angular velocity, is the robot joint angular acceleration, is the inertia matrix, and the joint angle Related, is the Coriolis force and centrifugal force matrix, and the joint angle and robot joint angular velocity Related, is the gravity vector, and the joint angle Related;

[0020] Substitute the designed control law, eliminate the nonlinear terms, and convert the designed Substitute into the original dynamic equation:

[0021] ;

[0022] Using the reversibility of the inertia matrix, decoupling is done into a linear system. Since the inertia matrix Is a positive definite matrix, multiply both sides by ,get: .

[0023] Further, step S04 specifically includes:

[0024] For a PCB using the PPC method, the upper bound of the performance constraint is expressed in the following two forms:

[0025] ;

[0026] Where, 、 、 、 is a positive real number designed, The time is calculated from the moment the robot system is started. It is a preset performance boundary function;

[0027] ;

[0028] Where, 、 、 、 is a positive real number of design, where The time is calculated from the moment the robot system is started. is the time it takes for the robot system to reach a steady state;

[0029] The conditions for steady-state or transient tracking performance meet the following formula:

[0030] ;

[0031] Where, , ( =1,2…n, represents the robot's Joint) represents the error between the actual angle reached by each joint of the robot and the expected angle reached; is the desired angle of the robot, The angle fed back in real time by the robot joints.

[0032] Furthermore, step S04 further includes:

[0033] A sliding preset performance control method is constructed based on PPC. The sliding preset performance control method satisfies the following constraints:

[0034] ;

[0035] in, and Representing upper SPCB and lower SPCB respectively, the expression is as follows:

[0036] ;

[0037] in, Represents SPCB, where:

[0038] ;

[0039] ;

[0040] pass The sign of the switch vector direction, adapting to different initial errors, where Represents the identity matrix, ensuring that the preset performance boundaries are adjusted synchronously in different directions;

[0041] ;

[0042] matrix Control the symmetry of preset performance boundaries;

[0043] in, represents the following displacement function:

[0044] ;

[0045] in, 、 is a positive real number, representing time, are design parameters, among which, The time is calculated from the moment the robot system is started. The time it takes for the robot system to reach a steady state.

[0046] Furthermore, the sliding preset performance control input Pick:

[0047] ;

[0048] in, , ,in 、 is an adjustable parameter. represents the feedback gain coefficient, represents the desired joint angular acceleration, ( =1,2…n, represents the robot's Joint) represents the error between the actual angle reached by each joint of the robot and the expected angle reached. It is a preset performance boundary function.

[0049] Furthermore, the torque formula of the joint in step S05 is:

[0050] ;

[0051] Where, is the robot joint angle, is the robot joint angular velocity, represents the desired joint angular acceleration, is the inertia matrix, and the joint angle Related, is the Coriolis force and centrifugal force matrix, and the joint angle and robot joint angular velocity Related, is the gravity vector, and the joint angle Related.

[0052] The present invention also discloses a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned sliding preset performance control method is executed.

[0053] The present invention also discloses a control system for a picking robot, the control system comprising:

[0054] The picking target determination module is used to obtain the determined position of the picking target as the target position of the robot during the picking process;

[0055] The robot end torque determination module is used to determine the torque information applicable to different picking targets;

[0056] The picking path planning module is used to plan the path for the robot to reach the picking target;

[0057] The control module is used to execute the above-mentioned picking robot sliding preset performance control method to drive the robot to complete picking according to the path.

[0058] The beneficial effects of the present invention are:

[0059] First, the control system of the picking robot of the present invention includes a control module and a picking target determination module. The control module can obtain the desired position and the desired posture of the robot end according to the picking target determination module, and calculate the desired absolute state parameters of each joint of the robot according to the robot inverse kinematics model. Then, the control module controls the robot to drive the end effector to move to the desired position according to the desired absolute state parameters of each joint of the robot.

[0060] Second, the control system of the harvesting robot of the present invention includes a robot end torque determination module, which senses the torque required for different targets through the torque sensor of the robot end effector to avoid damaging the target;

[0061] Third, the control system of the picking robot of the present invention also includes a picking path planning module, which can provide a picking path consisting of multiple position points, and then take each position point on the picking path as the desired position. The control module continuously controls the robot end effector to move along the position point on the picking path, thereby controlling the robot to move along the picking path. The control module also controls the speed of the joint to achieve the desired speed of the robot end effector to ensure smooth operation of the end, and further reduce the occurrence of jitter. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a flow chart of the sliding preset performance control method of the picking robot of the present invention;

[0063] Figure 2 This is a block diagram of the algorithm structure of the sliding preset performance control method of the picking robot of the present invention;

[0064] Figure 3 This is a tracking comparison diagram of a joint position of the picking robot sliding preset performance control method of the present invention;

[0065] Figure 4 This is the torque optimization diagram of the sliding preset performance control method of the picking robot of the present invention. DETAILED DESCRIPTION

[0066] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0067] like Figures 1 to 4 As shown in Example 1, a method for controlling the sliding preset performance of a picking robot includes the following steps:

[0068] Step S01: obtaining the target position of the robot during the picking process in real time;

[0069] Step S02: Acquire the position information, speed information and torque information of the robot joints;

[0070] Step S03: Designing a control law to offset the errors of gravity, Coriolis force and centrifugal force;

[0071] Step S04: introducing sliding preset performance control (SPPC) into the robot model, combined with the control law, to compensate for the error of the robot model calculated using the estimated inertial parameters;

[0072] Step S05: Optimize the torque input of the joints during the robot picking process based on the error and the error compensation result.

[0073] Example 2: This example is a further improvement on Example 1, and its details are as follows:

[0074] Step S01 specifically includes:

[0075] The target position is captured by the picking target determination module, and the expected parameters of each joint of the robot are calculated through the robot inverse kinematics model. Each joint moves according to the expected parameters to form an expected trajectory, so that the end of the robot reaches the expected posture.

[0076] Example 3: This example is a further improvement on Example 1, and its details are as follows:

[0077] The expression of the control law designed in step S03 is:

[0078] ;

[0079] Where, is the robot joint angle (position), is the robot joint angular velocity, is the inertia matrix (positive definite, reversible), and the joint angle (position) Related, is the Coriolis force and centrifugal force matrix, and the joint angle (position) and robot joint angular velocity Related, is the gravity vector, and the joint angle (position) Related, is the joint torque (control input), Slide preset performance control input for the design.

[0080] Example 4: This example is a further improvement on Example 3, and its details are as follows:

[0081] The expression of the robot model in step S04 is:

[0082] ;

[0083] Where, is the robot joint angle (position), is the robot joint angular velocity, is the robot joint angular acceleration, is the inertia matrix, and the joint angle (position) Related, is the Coriolis force and centrifugal force matrix, and the joint angle (position) and robot joint angular velocity Related, is the gravity vector, and the joint angle (position) Related;

[0084] Substitute the designed control law, eliminate the nonlinear terms, and convert the designed Substitute into the original dynamic equation:

[0085] ;

[0086] Using the reversibility of the inertia matrix, decoupling is done into a linear system. Since the inertia matrix Is a positive definite matrix (the physical meaning of the inertia matrix determines its reversibility), and both sides are multiplied by (inverse matrix), we get: , thereby actively counteracting the effects of gravity, Coriolis force and centrifugal force through the control law, and transforming the complex nonlinear system into a decoupled linear system (the acceleration of each joint Directly from control input Decide, =1,2…n, represents the robot's joints), and decouple the robot dynamics into the simplest linear relationship by offsetting the nonlinearity through the control law. , making complex robotic systems solvable and controllable.

[0087] Example 5: This example is a further improvement on Example 1, and its details are as follows:

[0088] Step S04 specifically includes:

[0089] PCB (performance bounds) using the PPC (preset performance control) method, where the upper bound of the performance constraint is expressed in the following two forms:

[0090] ;

[0091] Where, 、 、 、 is a positive real number designed, The time is calculated from the moment the robot system is started; It is a preset performance boundary function;

[0092] ;

[0093] Where, 、 、 、 is a positive real number of design, where The time is calculated from the moment the robot system is started. is the time when the robot system reaches a steady state (that is, the moment when the position error of each joint of the robot enters the preset steady state range and no longer changes significantly); In [0, ) interval (i.e. the transient stage from system startup to reaching steady state), (Preset performance boundary function) uses the expression ,when In [ ,+∞) interval (i.e. the stage after the system reaches steady state), Using expressions ; Ensure that the robot end reaches the target position within the specified time;

[0094] The conditions for steady-state or transient tracking performance meet the following formula:

[0095] ;

[0096] Where, , ( =1,2…n, represents the robot's Joint) represents the error between the actual angle (position) of each joint of the robot and the expected angle (position); is the desired angle (position) of the robot, The angle (position) fed back by the robot joint in real time.

[0097] Example 6: This example is a further improvement on Example 5, and its details are as follows:

[0098] If only the common PCB (performance constraint boundary) is used, the system's tracking error will experience significant jitter before reaching steady state. Furthermore, once the parameters are selected, the PCB (performance constraint boundary) is fixed. When the initial state (or reference signal) changes, it is necessary to re-verify whether the initial error still meets the initial constraint conditions. Therefore, a sliding preset performance control method (SPPC) is constructed to overcome these two limitations. This ensures that the initial performance constraint boundary can adaptively slide with the initial error, making it applicable to robotic systems with arbitrary bounded initial errors without degrading the control performance of the robot in the initial state.

[0099] A sliding preset performance control method is constructed based on PPC (preset performance control method). The sliding preset performance control method meets the following constraints:

[0100] ;

[0101] in, and They represent the upper SPCB (sliding performance constraint boundary) and the lower SPCB (sliding performance constraint boundary), respectively, and are expressed as follows:

[0102] ;

[0103] in, represents SPCB (Sliding Performance Constraint Boundary), where:

[0104] ;

[0105] ;

[0106] pass The sign of the switch vector direction, adapting to different initial errors, where Represents the identity matrix, ensuring that the preset performance boundaries are adjusted synchronously in different directions;

[0107] ;

[0108] matrix Control the symmetry of preset performance boundaries;

[0109] in, represents the following displacement function:

[0110] ;

[0111] in, 、 is a positive real number, representing time, are design parameters, among which, The time is calculated from the moment the robot system is started. is the time when the robot system reaches a steady state (that is, the moment when the position error of each joint of the robot enters the preset steady state range and no longer changes significantly); In [0, ) interval (i.e. the transient stage from system startup to reaching steady state), Using expressions ,when In [ ,+∞) interval (i.e. the stage after the system reaches steady state), Using expression 0; the above process enables the preset performance boundary to slide adaptively to accommodate any bounded initial error.

[0112] Example 7: This example is a further improvement on Example 6, and its details are as follows:

[0113] Designed sliding preset performance control input Pick:

[0114] ;

[0115] in, , , The bounded error is mapped to the unbounded error, so that the control rate can be sensitive to the boundary risk and accurately adjust the constraint strength of the preset performance boundary; and the bounded function can be used to and The nonlinear characteristics of the system ensure that the system is safe and operates according to the expected performance, 、 is an adjustable parameter. represents the feedback gain coefficient, represents the desired joint angular acceleration, ( =1,2…n, represents the robot's Joint) represents the error between the actual angle (position) of each joint of the robot and the expected angle (position). is a preset performance boundary function. This allows for compensation of robot model errors calculated using estimated inertial parameters.

[0116] Example 8: This example is a further improvement on Example 1, and its details are as follows:

[0117] The torque formula of the joint in step S05 is:

[0118] ;

[0119] Where, is the robot joint angle, is the robot joint angular velocity, represents the desired joint angular acceleration, is the inertia matrix, and the joint angle Related, is the Coriolis force and centrifugal force matrix, and the joint angle and robot joint angular velocity Related, is the gravity vector, and the joint angle Related; In this way, the torque of the robot can be controlled using the sliding preset performance control method;

[0120] Based on the real-time calculation of the robot's parameter error, the robot's output torque is optimized in real time to achieve precise picking results.

[0121] Embodiment 9, a computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the sliding preset performance control method of any one of embodiments 1 to 8 is executed.

[0122] Example 10, a control system for a picking robot, the control system comprising:

[0123] The picking target determination module is used to obtain the determined position of the picking target as the target position of the robot during the picking process;

[0124] The robot end torque determination module is used to determine the torque information applicable to different picking targets to avoid damaging the targets;

[0125] The picking path planning module is used to plan the path for the robot to reach the picking target;

[0126] The control module is used to execute the sliding preset performance control method of the picking robot in any one of embodiments 1 to 8, and drive the robot to complete picking according to the path.

[0127] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0128] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk, or an optical disk.

[0129] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A method for controlling the sliding preset performance of a picking robot, characterized in that: The steps include: Step S01: obtaining the target position of the robot during the picking process in real time; Step S02: Acquire the position information, speed information and torque information of the robot joints; Step S03: Designing a control law to offset the errors of gravity, Coriolis force and centrifugal force; Step S04: introducing a sliding preset performance control to the robot model, and combining the control law to compensate for the error of the robot model calculated using the estimated inertia parameters; Step S05: Optimizing the torque input of the joints during the picking process of the robot according to the error and the compensation result of the error; The step S04 includes: A sliding preset performance control method is constructed based on the preset performance control method. The sliding preset performance control method satisfies the following constraints: ； in, and They represent the upper sliding performance constraint boundary and the lower sliding performance constraint boundary respectively, and are expressed as follows: ； in, represents the sliding performance constraint boundary, where: ； ； Where, It is a preset performance boundary function; ( =1,2…n, represents the robot's Joint) represents the error between the actual angle reached by each joint of the robot and the expected angle reached; pass The sign of the switch vector direction, adapting to different initial errors, where Represents the identity matrix, ensuring that the preset performance boundaries are adjusted synchronously in different directions; ； matrix Control the symmetry of preset performance boundaries; in, represents the following displacement function: ； in, 、 is a positive real number, representing time, are design parameters, among which, The time is calculated from the moment the robot system is started. The time it takes for the robot system to reach a steady state.

2. A method for controlling sliding preset performance of a picking robot according to claim 1, characterized in that: The step S01 specifically includes: The target position is captured by the picking target determination module, and the expected parameters of each joint of the robot are calculated through the robot inverse kinematics model. Each joint moves according to the expected parameters to form an expected trajectory, so that the end of the robot reaches the expected posture.

3. The method for controlling sliding preset performance of a picking robot according to claim 1, characterized in that: The expression of the control law designed in step S03 is: ； Where, is the robot joint angle, is the robot joint angular velocity, is the inertia matrix, and the joint angle Related, is the Coriolis force and centrifugal force matrix, and the joint angle and robot joint angular velocity Related, is the gravity vector, and the joint angle Related, is the joint torque, Slide preset performance control input for the design.

4. A method for controlling sliding preset performance of a picking robot according to claim 3, characterized in that: The expression of the robot model in step S04 is: ； Where, is the robot joint angle, is the robot joint angular velocity, is the robot joint angular acceleration, is the inertia matrix, and the joint angle Related, is the Coriolis force and centrifugal force matrix, and the joint angle and robot joint angular velocity Related, is the gravity vector, and the joint angle Related; Substitute the designed control law, eliminate the nonlinear terms, and convert the designed Substitute into the original dynamic equation: ； Using the reversibility of the inertia matrix, decoupling is done into a linear system. Since the inertia matrix Is a positive definite matrix, multiply both sides by ,get: .

5. The method for controlling sliding preset performance of a picking robot according to claim 1, characterized in that: The step S04 specifically includes: The performance constraint boundary of the preset performance control method is adopted, where the upper boundary of the performance constraint is expressed in the following two forms: ； Where, 、 、 、 is a positive real number designed, The time is calculated from the moment the robot system is started. It is a preset performance boundary function; ； Where, 、 、 、 is a positive real number of design, where The time is calculated from the moment the robot system is started. is the time it takes for the robot system to reach a steady state; The conditions for steady-state or transient tracking performance meet the following formula: ； Where, , ( =1,2…n, represents the robot's Joint) represents the error between the actual angle reached by each joint of the robot and the expected angle reached; is the desired angle of the robot, The angle fed back in real time by the robot joints.

6. The method for controlling sliding preset performance of a picking robot according to claim 3, characterized in that: The design's sliding preset performance control input Pick: ； in, , ,in 、 is an adjustable parameter. represents the feedback gain coefficient, represents the desired joint angular acceleration, ( =1,2…n, represents the robot's Joint) represents the error between the actual angle reached by each joint of the robot and the expected angle reached. It is a preset performance boundary function.

7. A method for controlling sliding preset performance of a picking robot according to claim 6, characterized in that: The torque formula of the joint in step S05 is: ； Where, is the robot joint angle, is the robot joint angular velocity, represents the desired joint angular acceleration, is the inertia matrix, and the joint angle Related, is the Coriolis force and centrifugal force matrix, and the joint angle and robot joint angular velocity Related, is the gravity vector, and the joint angle Related.

8. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by a processor, the sliding preset performance control method according to any one of claims 1 to 7 is executed.

9. A control system for a picking robot, characterized in that: The control system includes: The picking target determination module is used to obtain the determined position of the picking target as the target position of the robot during the picking process; The robot end torque determination module is used to determine the torque information applicable to different picking targets; The picking path planning module is used to plan the path for the robot to reach the picking target; A control module is used to execute the sliding preset performance control method of the picking robot according to any one of claims 1 to 7, and drive the robot to complete picking according to the path.

Images