Tendon rope driven special operation robot joint position tracking time delay control method
By constructing a forgetting factor and a residual delay estimation error prediction term to compensate for the joint position tracking delay of the tendon-driven special operation robot, the problems of low accuracy and overshoot in the existing delay control methods are solved, and high-precision and overshoot-free joint position tracking is achieved.
Patent Information
- Application Number
- CN202511701465.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-19
- Publication Date
- 2026-02-27
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing time delay control methods cannot effectively solve the accuracy problem of joint position tracking in chord-driven special operation robots, especially when tracking time-varying reference signals, the accuracy is low and overshoot occurs when there are large step signals.
A forgetting factor calculation model, a residual time delay estimation error calculation model, and a time delay estimation error compensation model are constructed. The time delay estimation error is compensated by the forgetting factor and the residual estimation error prediction term. An improved time delay control law is constructed to achieve high-precision tracking.
It achieves high-precision tracking of joint positions in chord-driven special-purpose robots, maintaining high precision and no overshoot, especially with time-varying signals, and is unaffected by elastic stretching of the chords and measurement errors in the joint rotation radius.
Smart Images
Figure CN121572289A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of robot control technology, specifically relating to a method for controlling the time delay of joint position tracking in a chord-driven special-operation robot. Background Technology
[0002] Cable-driven robots are playing an increasingly important role in various specialized fields such as medical, aerospace, nuclear industry, and deep-sea operations. Cable drive is a driving method that uses tendons or flexible ropes to transmit power, offering advantages such as strong spatial adaptability and long transmission distance. Linear actuators such as electric cylinders or motors are typically located far from the end effector, and joint movement is achieved by pulling the tendons. The complex and delicate operations of cable-driven specialized robots rely heavily on high-precision joint position tracking. However, tendons are subject to elastic stretching and wear, which can lead to inaccuracies in the correspondence between joint angles and electric cylinder displacements. Furthermore, the measurement of joint rotation radius often contains errors, further compromising joint position tracking accuracy. Therefore, designing a joint position tracking control algorithm for cable-driven specialized robots that is independent of joint kinematic parameters is of great significance.
[0003] Time-delay control algorithms are widely used in robotics due to their low model dependency and computational simplicity; however, existing time-delay control methods cannot be directly applied to joint position tracking control of chord-driven special-task robots. Traditional time-delay control methods inherently suffer from time-delay estimation errors, which severely impact joint position tracking performance. To overcome this challenge, indirect compensation and direct compensation time-delay control methods have been proposed. An example of an indirect compensation time-delay control method is shown in the reference [J. Park, W. Kwon and P. Park. An improved adaptivesliding mode control based on time-delay control for robot manipulators]. IEEE Trans. Ind. Electron.[vol. 70, no. 10, pp. 10363-10373, Oct. 2023] aims to make the compensated time delay estimate approximate the true value of the uncertainty term, thereby reducing the joint angle tracking error. Although this improved method improves the joint position tracking accuracy to some extent, its tracking accuracy for time-varying reference signals (such as sine signals) is very limited. Another improved method is the direct compensation time delay control method, such as the reference [G. Wang, Y. Cao, H. Li and T. Shi. Discrete time delay control for current tracking of jointmotors with estimation error compensation]. IEEE Transactions on Industry Applications, PP The proposed method aims to minimize the compensated residual time delay estimation error, thereby converging the joint angle tracking error to zero via ideal error dynamics. While this method maintains good tracking accuracy for time-varying reference angle signals, it exhibits significant angle overshoot when tracking large-amplitude step reference signals. Therefore, there is an urgent need to propose an improved time delay control method that can simultaneously achieve high-precision tracking of time-varying signals and overshoot-free tracking of large-amplitude step signals. Summary of the Invention
[0004] In view of the above, the present invention provides a method for joint position tracking delay control of a chord-driven special operation robot, which can compensate for delay estimation error and reduce joint position tracking error.
[0005] A method for controlling the time delay of joint position tracking in a chord-driven special-purpose robot includes the following steps: (1) Establish a dynamic mapping model between the joint angles and the displacement extension of the electric cylinder (linear actuator) of the special operation robot driven by tendon rope; (2) Construct a forgetting factor calculation model and calculate the forgetting factor at the current moment based on the difference between the reference joint angle and the actual joint angle of the robot at the previous moment. (3) Construct a model for calculating the remaining time delay estimation error. Calculate the remaining time delay estimation error at the current moment based on the difference between the robot's reference joint angle and the actual joint angle at the previous moment and its first derivative. (4) Construct a prediction model for the remaining time delay estimation error, and calculate the remaining time delay estimation error for the next time step based on the remaining time delay estimation error values of the previous and current time steps and the forgetting factor of the current time step; (5) Construct a time delay estimation error compensation model. Calculate the time delay estimation error compensation value at the current time based on the time delay estimation error compensation value at the previous time, the forgetting factor at the current time, and the predicted value of the remaining time delay estimation error at the next time. (6) Construct an improved time delay control law. Based on the time delay estimate at the current moment, the difference between the reference joint angle and the actual joint angle of the robot at the previous moment and its first derivative, calculate the output of the time delay control law at the current moment. The output is the displacement extension amount applied to the electric cylinder at the current moment. (7) Calculate the displacement reference value of the electric cylinder based on the displacement extension and retraction amount applied to the electric cylinder at the current moment, and then control it sequentially through the position loop, speed loop and current loop to make the electric cylinder extend and retract to the given position and drive the tendon rope to move, thereby pulling the joint to rotate the corresponding angle, and realizing high-precision tracking of the joint position of the tendon rope driven special operation robot.
[0006] Furthermore, the expression for the dynamic mapping model in step (1) is as follows: in: Indicates the first k The amount of displacement or extension applied to the electric cylinder at all times. Indicates the first k The actual joint angles of the robot at all times. for The first-order differential, Indicates the first k The total uncertainty that constantly affects the displacement and contraction of the electric cylinder (including length deviations caused by factors such as the elastic stretching of the tendon ligament and the fine adjustment of the spatial position of the tendon ligament, as well as the calculated length deviations caused by the measurement error of the joint rotation radius), A The gain coefficient is a constant. T s To control the cycle, k It is a natural number.
[0007] Furthermore, the expression for the forgetting factor calculation model in step (2) is as follows: in: Indicates the first k The forgetting factor of time, This represents the natural exponential function. Indicates the first k The difference between the robot's reference joint angle and the actual joint angle at time -1. Indicates the first k The robot's reference joint angle at time -1. Indicates the first kThe robot's actual joint angle at time -1.
[0008] Furthermore, the expression for the residual delay estimation error calculation model in step (3) is as follows: in: Indicates the compensation The remaining time delay estimation error, for The first-order differential, K p is the convergence coefficient.
[0009] Furthermore, the expression for the remaining time delay estimation error prediction model in step (4) is as follows: in: Indicates the first k The predicted value of the remaining time delay estimation error at time +1. Indicates the compensation The remaining time delay estimation error, Indicates the first k -1 is the lumped uncertainty term affecting the displacement contraction of the electric cylinder.
[0010] Furthermore, the expression for the time delay estimation error compensation model in step (5) is as follows: in: and The first k Time and the k The time delay estimation error compensation value at time -1. K c To adjust the step size.
[0011] Furthermore, the expression for the time delay control law in step (6) is as follows: in: For the first k Injection terms of the time delay control law, for The estimated time delay.
[0012] Furthermore, the expression for the injected item is as follows: in: for The first differential.
[0013] Furthermore, the time delay estimate The expression is as follows: in: Indicates the first k The displacement / extension applied to the electric cylinder at time -1 for The first-order differential, Indicates the first k The robot's actual joint angle at time -1.
[0014] Furthermore, in step (7), the displacement reference value of the electric cylinder is calculated using the following formula: in: Indicates the first k The reference value of the electric cylinder's displacement at all times. This indicates the maximum displacement of the electric cylinder.
[0015] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described method for joint position tracking delay control of a tendon-driven special-operation robot.
[0016] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the aforementioned method for joint position tracking delay control of a chord-driven special-operation robot.
[0017] Based on the above technical solution, the present invention has the following beneficial technical effects: 1. This invention addresses the problem of low tracking accuracy caused by inherent time delay estimation errors in traditional time delay control algorithms. It compensates for the time delay estimation errors by adding a compensation term to the control law of the improved time delay control algorithm, thereby achieving higher accuracy in joint position tracking of chord-driven special operation robots.
[0018] 2. This invention addresses the low precision problem of indirect compensation time delay control algorithms when tracking time-varying reference signals and the large overshoot problem of direct compensation time delay control algorithms when tracking large-amplitude step reference signals. It introduces a forgetting factor and a residual estimation error prediction term in the design of the time delay estimation error compensation term, thereby achieving high-precision tracking of time-varying signals and overshoot-free tracking of large-amplitude step signals.
[0019] 3. The time delay control algorithm used in this invention can achieve high-precision tracking of the joint position of chord-driven special operation robots, and is not affected by factors such as chord elastic stretching, chord spatial position fine adjustment, and joint rotation radius measurement errors.
[0020] 4. This invention verifies the robustness of the time delay control algorithm used in this invention by testing it on the joints of special-purpose robots driven by different tendon cables and by performing position tracking tests under reference commands at different positions.
[0021] 5. The control law used in this invention can achieve high-precision position tracking control for single joints, and can also be extended to multi-joint applications to achieve high-precision position tracking control for multiple joints of chord-driven special operation robots.
[0022] Therefore, this invention can compensate for time delay estimation errors. By adding a compensation term to the control law, it reduces the joint position tracking error of the chord-driven special operation robot. By introducing a forgetting factor and a residual estimation error prediction term into the compensation term, it can simultaneously achieve high-precision tracking of time-varying signals and overshoot-free tracking of large-amplitude step signals. At the same time, it is not affected by factors such as chord elastic tension, chord spatial position fine adjustment, and joint rotation radius measurement errors, and has strong robustness and scalability. Attached Figure Description
[0023] Figure 1 This is an overall control block diagram of the joint position tracking delay control system for the tendon-driven special operation robot of the present invention.
[0024] Figure 2 This is a flowchart illustrating the joint position tracking delay control method for a chord-driven special-operation robot according to the present invention.
[0025] Figure 3 Figure 1 shows the experimental results of joint position tracking under large-amplitude step reference joint angle signal conditions, comparing the traditional time delay control algorithm with the improved time delay control algorithm of this invention. In the figure, (a) shows the experimental results under a 50-degree step reference joint angle signal, and (b) shows the experimental results under a 30-degree step reference joint angle signal.
[0026] Figure 4 Figure 1 shows the experimental results of joint position tracking under time-varying (sinusoidal) reference angle signal conditions, using the traditional time delay control algorithm and the improved time delay control algorithm of this invention. In the figure, (a) shows the experimental results under a sinusoidal reference angle signal with an amplitude of 30 degrees, and (b) shows the experimental results under a sinusoidal reference angle signal with an amplitude of 20 degrees. Detailed Implementation
[0027] To describe the present invention in more detail, the technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
[0028] like Figure 1As shown, the joint position tracking delay control system of the tendon-wire driven special operation robot of the present invention includes key links such as forgetting factor calculation, remaining delay estimation error calculation, remaining delay estimation error prediction, delay estimation error compensation value calculation, ideal error dynamic equation construction, delay estimation, and delay control algorithm control output. Among them, the forgetting factor calculation, remaining delay estimation error prediction, and delay estimation error compensation value calculation are key links that distinguish it from traditional delay control algorithms.
[0029] For the forgetting factor calculation, the reference joint angle of the chord-driven special-task robot at the previous moment is used. and actual joint angle The input forgetting factor calculation model is used to calculate the forgetting factor at the current time, and the output is the forgetting factor at the current time after the natural exponential function is applied. The specific generation formula is as follows: For the calculation of the remaining time delay estimation error, the remaining time delay estimation error at the current time... Based on the angle tracking error of the previous moment and its first differential The calculation is as follows: The angle tracking error of the joints of the chord-driven special operation robot at the previous moment. Reference angle from the previous moment From a practical perspective The difference is expressed as: For the remaining time delay estimation error prediction stage, the predicted value of the remaining time delay estimation error prediction term at the next time step... The error value is estimated from the remaining time delay of the previous time and the current time. and And the forgetting factor at the current moment The calculated prediction formula is expressed as follows: For the calculation of the time delay estimation error compensation value, the time delay estimation error compensation value at the current moment... Error compensation value based on the time delay estimation of the previous moment Forgetting factor at the current moment β ( kT s And the predicted value of the remaining time delay estimation error at the next moment. The calculation formula for the time delay estimation error compensation value is as follows: For the construction of the ideal error dynamics equation, the actual joint angle of the chord-driven special-operation robot at the previous moment... Reference joint angle and its first derivative value Input the ideal error dynamics equation construction module and output the injection term at the current time. The specific formula for generating this formula is as follows: For the time delay estimation stage, the lumped uncertainty term of the chord-driven special-operation robot at the current moment... Delay estimate The amount of extension / retraction of the electric cylinder at the previous moment First-order differential values of actual joint angles in chord-driven special-operation robots The time delay estimation equation is obtained through time delay estimation: The control law of the traditional time delay control algorithm can be obtained from the above ideal error dynamics equation construction stage and time delay estimation stage, as follows: The control law of the above traditional time delay control algorithm is used for joint position tracking of a chord-driven special operation robot, and the experimental results of its joint position tracking are used to compare with the time delay control algorithm of the present invention.
[0030] For the control output stage of the time-delay control algorithm, a time-delay estimation error compensation value is added to the traditional time-delay control algorithm control law, which controls the output of the chord-driven special operation robot at the current moment. Error compensation value based on the time delay estimation at the current moment Control Law of Traditional Time Delay Control Algorithm The following calculations were performed: The control output of the obtained time delay control algorithm In practical terms, it refers to the displacement and extension of the electric cylinder, and the reference value of the electric cylinder's displacement at the current moment. The maximum displacement of the electric cylinder And the displacement and extension of the electric cylinder The calculation is as follows: The obtained electric cylinder displacement reference value serves as the input to the built-in motor position loop. After proportional control of the position loop, it outputs the input to the motor speed loop. The speed loop, through proportional-integral control, outputs the input to the motor current loop. The current loop, also through proportional-integral control, outputs the required drive voltage. This drive voltage, after coordinate transformation and space vector pulse width modulation, controls the switching devices of the inverter circuit to turn on and off, thereby driving the motor to rotate and causing the electric cylinder to extend and retract to a given position. The extension and retraction of the electric cylinder drives the tendon cable to move, thereby pulling the joint to rotate by the corresponding angle, achieving high-precision tracking of the joint position of the tendon cable-driven special-purpose robot. The actual joint angle of the tendon cable-driven special-purpose robot is detected by sensors and fed back to the controller for calculating the control law for the next moment.
[0031] In this embodiment, the control unit is composed of an STM32 control unit with a control frequency set to 1kHz. An encoder is integrated inside the joint for measuring the actual joint angle. The measured value is fed back to the time delay control law, which outputs the displacement extension of the electric cylinder and sends it to the electric cylinder. The electric cylinder integrates a controller and a driver to run position loop, speed loop, and current loop control programs. During the experiment, data such as joint angle and electric cylinder displacement are transmitted to the host computer via CAN box and recorded.
[0032] The specific implementation process of the tendon-wire driven special operation robot joint position tracking time delay control method of the present invention is as follows: Figure 2 As shown: First, a dynamic mapping model is established between the joint angles of the chord-driven special-purpose robot and the displacement and extension of the electric cylinder. The geometric relationship between the joint angles of the chord-driven special-purpose robot and the length of the driving chord at the previous and current moments is as follows: in: Indicates the length of the driving tendon chord. k and T s These represent the time and control period in the discrete time domain, respectively. R This indicates the radius of rotation of the joints in a chord-driven special-purpose robot. This indicates the actual joint angles of a chord-driven special-purpose robot.
[0033] The relationship between the electric cylinder displacement and the length of the drive tendon rope at the previous moment and the current moment is as follows: in: This indicates the displacement and extension amount of the electric cylinder. This indicates the maximum displacement of the electric cylinder. This indicates the length deviation caused by factors such as the elastic stretching of the tendon ligament and minor adjustments to its spatial position.
[0034] The relationship between the joint angle and the displacement / extension of the electric cylinder is derived from the relationship between the joint angle and the length of the driving tendon ligament, and the relationship between the electric cylinder displacement and the length of the driving tendon ligament, as follows: Meanwhile, the first-order differential value of the actual joint angle of the chord-driven special operation robot at the current moment The actual joint angles of the chord-driven special operation robot at the current moment The actual joint angles of the chord-driven special operation robot at the previous moment The following relationship exists between them: Therefore, the relationship between the joint angle and the displacement / extension of the electric cylinder can be further written as: Since the measurement of joint rotation radius is often inaccurate, a constant coefficient is introduced by human intervention. A The relationship between the joint angle and the displacement / extension of the electric cylinder can be rewritten as follows: Lumped uncertainty term affecting the displacement contraction of the electric cylinder This includes length deviations caused by factors such as elastic stretching of the tendon ligament and fine-tuning of its spatial position, as well as calculated length deviations caused by measurement errors in the joint rotation radius.
[0035] Thus, the dynamic mapping model between the joint angles and the displacement and extension of the electric cylinder of the chord-driven special operation robot is established. The time delay control method of this invention will be based on the above dynamic mapping model.
[0036] Then, the difference between the reference joint angle and the actual joint angle of the chord-driven special operation robot is calculated as follows: in: This indicates the joint position tracking error of a chord-driven special-purpose robot. This indicates the reference joint angle for chord-driven special-purpose robots.
[0037] A forgetting factor calculation model is constructed. Based on the difference between the reference joint angle and the actual joint angle of the chord-driven special-task robot at the previous moment, the forgetting factor value at the current moment is calculated. The calculation model of the forgetting factor is as follows: in: This represents the forgetting factor.
[0038] Next, a residual time delay estimation error calculation model is constructed. Based on the difference between the reference joint angle and the actual joint angle of the chord-driven special operation robot at the previous moment, and the first derivative of the angle tracking error, the residual time delay estimation error value at the current moment is calculated. The residual time delay estimation error calculation model is as follows: in: This represents the compensated uncertainty term of the aggregate. The remaining time delay estimation error, K p This represents the convergence parameter for joint position tracking error in chord-driven special-purpose robots.
[0039] Following this, a prediction model for the remaining time delay estimation error prediction term at the next time step is constructed. Based on the remaining time delay estimation error values of the previous and current time steps, as well as the forgetting factor value at the current time step, the predicted value of the remaining time delay estimation error at the next time step is calculated. The prediction model for the remaining time delay estimation error prediction term is as follows: in: This represents the predicted value of the remaining time delay estimation error at the next time step.
[0040] Subsequently, a time delay estimation error compensation model is constructed. Based on the time delay estimation error compensation value of the previous time step, the value of the forgetting factor at the current time step, and the predicted value of the remaining time delay estimation error at the next time step, the time delay estimation error compensation value at the current time step is calculated. The time delay estimation error compensation model is as follows: in: This represents the compensation value for the time delay estimation error. K c This indicates the correction step size for the time delay estimation error compensation value.
[0041] Then, an improved time-delay control algorithm control law is constructed. Based on the current time-delay estimate, the difference between the reference joint angle and the actual joint angle of the chord-driven special-operation robot at the previous moment, and the first derivative of the angle tracking error, the control law output of the time-delay control algorithm at the current moment is calculated to obtain the displacement extension / retraction amount to be applied to the electric cylinder. The improved time-delay control algorithm control law is as follows: in: This indicates the amount of displacement or extension that will be applied to the electric cylinder. This represents the injection item in the delay control algorithm. This represents the lumped uncertainty term affecting the displacement contraction of the electric cylinder. The estimated time delay, Indicates the reference joint angle of a chord-driven special-purpose robot. The first differential.
[0042] Finally, the reference displacement value of the electric cylinder is calculated based on the maximum displacement and the displacement extension / retraction of the electric cylinder. The calculation process is as follows: in: This indicates the reference value for the displacement of the electric cylinder. This indicates the maximum displacement of the electric cylinder.
[0043] The obtained electric cylinder displacement reference value As the input to the built-in motor position loop, the motor speed loop is output after proportional control of the position loop. The speed loop outputs the motor current loop through proportional-integral control. The current loop outputs the required drive voltage through proportional-integral control. This drive voltage, after coordinate transformation and space vector pulse width modulation, controls the switching devices of the inverter circuit to turn on and off, thereby driving the motor to rotate and causing the electric cylinder to extend and retract to a given position. The extension and retraction of the electric cylinder drives the tendon cable to move, thereby pulling the joint to rotate by the corresponding angle, realizing high-precision tracking of the joint position of the tendon cable-driven special operation robot.
[0044] In this embodiment, all constant coefficients in the delay control law expression are set to... A =0.0005, K p =10, K c =6000.
[0045] To verify the superiority of the improved time delay control algorithm of this invention, we compared the improved time delay control algorithm with the traditional time delay control algorithm through experiments, such as... Figure 3 The figure shows the joint position tracking experimental results of both under the condition of a large step reference joint angle signal. Figure 3In experiment (a), the specific operating conditions were set as follows: the joint reference angle was set to a step amplitude of 50 degrees, i.e., a step from 10 degrees to 60 degrees (positive step), and a step back from 60 degrees to 10 degrees (negative step). It is worth noting that the adjustment time in the experiment was recorded according to a 5% threshold. Experimental results show that for the traditional time-delay control algorithm, the overshoot for 50-degree positive step tracking is 2.54%, and the adjustment time is 3.85s; the overshoot for 50-degree reverse step tracking is 2.31%, and the adjustment time is 3.40s; while under the improved time-delay control algorithm of this invention, the overshoot for 50-degree positive step tracking is 2.37%, and the adjustment time is 1.02s; the overshoot for 50-degree reverse step tracking is 2.19%, and the adjustment time is 1.01s. For example... Figure 3 In experiment (b), the specific operating conditions were set as follows: the joint reference angle was set to a step amplitude of 30 degrees, i.e., a step from 10 degrees to 40 degrees (positive step), and a step back from 40 degrees to 10 degrees (negative step). Experimental results showed that under the traditional time-delay control algorithm, the overshoot for tracking a 30-degree positive step was 2.45%, and the settling time was 3.98 s; the overshoot for tracking a 30-degree negative step was 2.49%, and the settling time was 0.96 s. However, under the improved time-delay control algorithm of this invention, the overshoot for tracking a 30-degree positive step was 2.20%, and the settling time was 0.96 s; the overshoot for tracking a 30-degree negative step was 2.39%, and the settling time was 0.93 s. In tracking the step reference joint angle signals of the above two amplitudes, the tracking overshoot and settling time of the improved time-delay control algorithm of this invention were both lower than those of the traditional time-delay control algorithm. This is attributed to the rapidly changing forgetting factor, and the experimental results for the forgetting factor also show… Figure 3 The results demonstrate the effectiveness of the improved time delay control algorithm of this invention in achieving overshoot-free tracking when tracking a large step reference signal.
[0046] like Figure 4 The figure shows the experimental results of joint position tracking under time-varying (sinusoidal) reference angle signal conditions using the traditional time-delay control algorithm and the improved time-delay control algorithm of this invention. Figure 4 In (a) of the experiment, the specific operating conditions were set as follows: the time-varying joint angle reference signal was set to a sinusoidal command with an amplitude of 30 degrees, a frequency of 0.1 Hz, and an offset of 40 degrees. q ref =40+30sin(20 t Experimental results show that, under the control of the traditional time-delay control algorithm, the root mean square error of joint angle tracking is 12.81 degrees, and its reference input angle is... q ref Compared with the actual output angle qThe input-output characteristic curves between these curves approximate an ellipse; however, under the improved time-delay control algorithm of this invention, the root mean square error of joint angle tracking is reduced to 0.825 degrees, a decrease of 93.56%, compared to the reference input angle. q ref Compared with the actual output angle q The input-output characteristic curves between them are close to a straight line. For Figure 4 In (b) of the experiment, the specific operating conditions were set as follows: the time-varying joint angle reference signal was set to a sinusoidal command with an amplitude of 20 degrees, a frequency of 0.1 Hz, and an offset of 30 degrees. q ref =30+20sin(20 t Experimental results show that, under the control of the traditional time-delay control algorithm, the root mean square error of joint angle tracking is 10.27 degrees, and its reference input angle is... q ref Compared with the actual output angle q The input-output characteristic curves between these curves approximate an ellipse; however, under the improved time-delay control algorithm of this invention, the root mean square error of joint angle tracking is reduced to 0.640 degrees, a decrease of 93.77%, based on the reference input angle. q ref Compared with the actual output angle q The input-output characteristic curves are close to a straight line. Under time-varying reference joint angle conditions with different sinusoidal amplitudes, the root mean square error of joint angle tracking of the improved time delay control algorithm of this invention is much lower than that of the traditional time delay control algorithm. This is due to the almost constant forgetting factor, and the experimental results of the forgetting factor also show... Figure 4 The results demonstrate the effectiveness of the improved time delay control algorithm of this invention in high-precision current tracking under constant reference current and variable speed conditions.
[0047] The above description of the embodiments is provided to enable those skilled in the art to understand and apply the present invention. Those skilled in the art can readily make various modifications to the above embodiments and apply the general principles described herein to other embodiments without creative effort. Therefore, the present invention is not limited to the above embodiments, and any improvements and modifications made to the present invention by those skilled in the art based on the disclosure thereof should be within the scope of protection of the present invention.
Claims
1. A tendon drive special purpose robot joint position tracking latency control method, characterized by, The method comprises the following steps: (1) establishing a dynamic mapping model between the joint angle of the tendon-driven special operation robot and the displacement extension and contraction amount of the electric cylinder; (2) constructing a forgetting factor calculation model, and calculating the forgetting factor at the current moment according to the difference between the reference joint angle and the actual joint angle of the robot at the previous moment; (3) constructing a residual time delay estimation error calculation model, and calculating the residual time delay estimation error at the current moment according to the difference between the reference joint angle and the actual joint angle of the robot at the previous moment and the first-order differential thereof; (4) constructing a residual time delay estimation error prediction model, and calculating the residual time delay estimation error at the next moment according to the residual time delay estimation error values at the previous moment and the current moment and the forgetting factor at the current moment; (5) constructing a time delay estimation error compensation model, and calculating the time delay estimation error compensation value at the current moment according to the time delay estimation error compensation value at the previous moment, the forgetting factor at the current moment and the prediction value of the residual time delay estimation error at the next moment; (6) constructing an improved time delay control law, and calculating the output of the time delay control law at the current moment according to the time delay estimation value at the current moment, the difference between the reference joint angle and the actual joint angle of the robot at the previous moment and the first-order differential thereof, wherein the output is the displacement extension and contraction amount applied to the electric cylinder at the current moment; (7) calculating the displacement reference value of the electric cylinder according to the displacement extension and contraction amount applied to the electric cylinder at the current moment, and then sequentially controlling through the position loop, the speed loop and the current loop to make the electric cylinder extend and contract to the given position and drive the tendon to move, so as to pull the joint to rotate the corresponding angle, thereby realizing high-precision tracking of the joint position of the tendon-driven special operation robot.
2. The tendon drive special purpose robot joint position tracking latency control method of claim 1, wherein, The expression of the dynamic mapping model in the step (1) is as follows: wherein: denotes the displacement extension amount applied to the electric cylinder at the k time t, denotes the actual joint angle of the robot at the k time t, is the first derivative of denotes the lumped uncertainty term affecting the displacement contraction amount of the electric cylinder at the k time t, A is a constant gain coefficient, T s is a control period, k is a natural number. 3. The tendon drive special purpose robot joint position tracking latency control method of claim 2, wherein, The expression of the forgetting factor calculation model in the step (2) is as follows: wherein: denotes the reference joint angle of the robot at time k denotes the forgetting factor at time denotes the natural exponential function, denotes the reference joint angle of the robot at time k denotes the difference between the reference joint angle and the actual joint angle of the robot at time denotes the reference joint angle of the robot at time k denotes the reference joint angle of the robot at time denotes the actual joint angle of the robot at time k denotes the actual joint angle of the robot at time 4. The tendon drive special purpose robot joint position tracking latency control method of claim 3, wherein, The expression of the residual time delay estimation error calculation model in the step (3) is as follows: wherein: represents the compensated pair of residual delay estimation errors, is a first derivative of K p is a convergence coefficient.
5. The tendon drive special purpose robot joint position tracking latency control method of claim 4, wherein, The expression of the residual time delay estimation error prediction model in the step (4) is as follows: in: Indicates the first k The predicted value of the remaining time delay estimation error at time +1. Indicates the compensation The remaining time delay estimation error, Indicates the first k -1 is the lumped uncertainty term affecting the displacement contraction of the electric cylinder.
6. The tendon drive special purpose robot joint position tracking latency control method of claim 5, wherein, The expression of the time delay estimation error compensation model in the step (5) is as follows: wherein: and are the time delay estimation error compensation values at the k time instant and at the k -1 time instant, respectively, K c is the correction step.
7. The tendon drive special purpose robot joint position tracking latency control method of claim 6, wherein, The expression of the time delay control law in the step (6) is as follows: wherein: is the k is the is the is the time delay estimate.
8. The tendon drive special purpose robot joint position tracking latency control method of claim 7, wherein, The expression of the injection term is as follows: wherein: is the first derivative.
9. The tendon drive special purpose robot joint position tracking latency control method of claim 7, wherein, The time delay estimate The expression is as follows: wherein: represents the first k -1 time displacement expansion amount applied to the electric cylinder, is the first-order differential of represents the first k -1 time actual joint angle of the robot.
10. The tendon drive special purpose robot joint position tracking latency control method of claim 7, wherein, The displacement reference value of the electric cylinder is calculated through the following relationship in the step (7): wherein: represents the first k represents the displacement reference value of the electric cylinder at the time, represents the maximum displacement value of the electric cylinder.