A speed sensorless control method for joint memory motor of humanoid robot

Abstract

The invention discloses a speed sensorless control method for a humanoid robot joint memory motor, and relates to the technical field of electric drive control equipment. Firstly, by observing the d-axis stator current component and the q-axis stator current component of the humanoid robot joint memory motor, the rotation speed, electromagnetic torque, magnetic flux, and the d-axis stator voltage component and the q-axis stator voltage component are designed, and the relationship between the current and the electrical angle of the joint motor is established. Then, according to the equivalent control principle, the observation value of the electrical angle is obtained, and the sliding surface is defined by the observation value of the electrical angle. Through the second-order generalized integral sliding mode control strategy, a controller structure is designed, and a suitable adaptive sliding mode gain coefficient is added to ensure the accuracy of the control system, so that good performance can be achieved at different rotation speeds. Finally, according to the Lyapunov theorem, stability analysis is completed to prove that the estimated electrical angle observation value enables the humanoid robot joint memory motor to run smoothly.

Description

Technical Field

[0001] The invention relates to the technical field of electric drive control equipment, in particular to a speed sensorless control method of a joint memory motor of a humanoid robot. Background Art

[0002] In the transmission operation of the joint memory motor of the humanoid robot, the uncertainty of parameter changes causes the output value of the speed sensorless motor to fluctuate with external factors. In order to ensure the motion stability of the humanoid robot, it is necessary to evaluate the online identification parameters of the rotor electrical angular velocity in real time. The controller based on mechanical speed sensor will bring problems such as high cost, additional space occupation and low reliability. Therefore, speed sensorless control has attracted more and more attention. Speed ​​information monitoring is an important indicator for performance evaluation, so the design of joint memory motor sensors is particularly important. Compared with traditional permanent magnet motors, memory motors have the important feature of adjustable magnetic flux. They are the core components for realizing the electromechanical energy conversion of robot joint memory motors. They are in an important position of energy conversion and play the role of matching speed and transmitting torque between the prime mover and the working machine or actuator. They are of great significance to ensure the reliable operation of the robot.

[0003] After searching relevant patents and literature at home and abroad, the main speed sensorless control methods include PI adaptive controller method, extended Kalman filter method, model reference adaptive method, extended Kalman filter method, etc. Although there are many methods for improving the speed detection of robot joint memory motors, no patents and literature on speed sensorless control methods for humanoid robot joint memory motors have been found. Literature on speed sensorless control methods for related motors, such as "Speed ​​sensorless direct torque control method for double-fed induction motors", obtains the speed observation value of the generator by calculating the synchronous speed and slip speed of the rotor flux. And "Stability improvement strategy for low-speed power generation conditions of speed sensorless asynchronous motors considering parameter errors" For speed sensorless asynchronous motor control systems, the feedback matrix can theoretically reduce the unstable area of ​​system power generation operation and improve the speed observation performance under low-speed power generation conditions, but the above two control methods lack online identification capabilities for speed information detection under the disturbance of given signals and external loads. The existing speed sensorless control method for controlling motors is difficult to be directly applied to the joint memory motor system of humanoid robots due to the limitations of estimated speed and accuracy. Traditional speed sensorless control has problems such as slow response and large overshoot in the application of humanoid robot joint memory motor control. Summary of the invention

[0004] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and provide a speed sensorless control method for a humanoid robot joint memory motor. The present invention can ensure the accuracy of the control system, achieve good performance at different speeds, and ensure smooth operation of the humanoid robot joint memory motor.

[0005] The present invention adopts the following technical solutions to solve the above technical problems:

[0006] According to a speed sensorless control method of a humanoid robot joint memory motor proposed by the present invention,

[0007] Firstly, by observing the d-axis stator current component and the q-axis stator current component of the humanoid robot joint memory motor, the speed, electromagnetic torque, flux linkage, d-axis stator voltage component and q-axis stator voltage component are designed, and the relationship between the current and electrical angular velocity of the joint memory motor is established.

[0008] Then, the observed value of the electrical angular velocity is obtained according to the equivalent control principle, and the sliding surface is defined by the observed value of the electrical angular velocity. The controller structure is designed through the second-order generalized integral sliding mode control strategy, and an adaptive sliding mode gain coefficient is added.

[0009] Finally, the stability analysis was completed according to the Lyapunov theorem, and the estimated electrical angular velocity observations made the joint memory motors of the humanoid robot run smoothly.

[0010] As a further optimization scheme of the speed sensorless control method of the joint memory motor of a humanoid robot described in the present invention, a relationship between the current and the electrical angular velocity of the joint memory motor is established, which is as follows:

[0011] By observing the stator current component i of the d-axis of the humanoid robot joint memory motor d and the q-axis stator current component i q , design speed, electromagnetic torque, flux linkage and d-axis stator voltage component u d and the q-axis stator voltage component u q The dynamic mathematical model of the current is used to adjust the electrical angular velocity of the humanoid robot joint memory motor through the PI controller, and the d-axis stator current component i is obtained. d and the q-axis stator current component i q The control model is:

[0012]

[0013] Among them, ω c is the electrical angular velocity, t is the time, R is the resistance, L is the inductance, is u d The estimated value of is u q The estimated value of is i d The estimated value of is i q The estimated value of .

[0014] As a further optimization scheme of the speed sensorless control method of the humanoid robot joint memory motor described in the present invention, the observation value of the electrical angular velocity is obtained according to the equivalent control principle, the sliding surface is defined by the observation value of the electrical angular velocity, and the controller structure is designed through the second-order generalized integral sliding mode control strategy, and an adaptive sliding mode gain coefficient is added; the details are as follows:

[0015] According to the equivalent control principle, the electrical angular velocity ω is obtained c Observed value

[0016]

[0017] Among them, K p and K i are the two parameters involved in PI control, ψ f is the magnetic flux, given by the electrical angular velocity ω c Observed value Define the sliding surface s;

[0018]

[0019] Through the second-order generalized integral sliding mode control strategy, the controller structure is designed as follows:

[0020]

[0021] Among them, sign(·) is the sign function, u is the control rate, is the derivative of u, is the derivative of s; k v is the adaptive sliding mode gain coefficient;

[0022] In the speed regulation process of the humanoid robot joint memory motor, an adaptive sliding mode gain coefficient k is added. v Ensure the accuracy of the control memory motor system, as shown in formula (5),

[0023]

[0024] Among them, k min is the minimum sliding mode gain, k max is the highest sliding mode gain, l is the adaptive gain adjustment parameter, is i α The estimated value of α is the α-axis current, is i β The estimated value of βis the β-axis current;

[0025] According to the spatial structure of the sliding surface, the second-order sliding mode switching function limits the boundary of the memory motor system's motion by narrowing the convergence range of the memory motor system.

[0026] As a further optimization scheme of the speed sensorless control method of the humanoid robot joint memory motor described in the present invention, stability analysis is completed according to the Lyapunov theorem, and the estimated electrical angular velocity observation value enables the humanoid robot joint memory motor to run smoothly; specifically as follows:

[0027] In a sampling period, the state vector matrix η is selected and the stability analysis is completed according to the Lyapunov theorem.

[0028]

[0029] Where t is time and superscript T is transpose;

[0030] If the designed adaptive sliding mode gain coefficient k v When the sliding surface s is reached in a finite time, the derivative of the state vector matrix of the humanoid robot joint memory motor system is Converges to 0 in a finite time;

[0031]

[0032] Where P is the state matrix of the memory motor system;

[0033] Select the electrical angular velocity ω c Observed value And the derivative of formula (7) is:

[0034]

[0035] Among them, η T is the transposed matrix of η, yes The transposed matrix of

[0036] The stability of η is determined. When η is a positive definite matrix, the matrix is a negative definite matrix of the state vector matrix η, then The positive definite condition is satisfied, and the memory motor system is stable at the equilibrium point, that is, the designed second-order generalized integral sliding mode control strategy is stable, which means that the estimated electrical angular velocity observation value Make the humanoid robot joint memory motor run smoothly.

[0037] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0038] (1) Compared with the traditional speed sensorless control system, the speed sensorless control method based on the second-order generalized integral sliding mode in the present invention can improve the anti-interference ability of the humanoid robot joint memory motor system, significantly reduce the overshoot and effectively solve the chattering problem, so that the speed sensorless control system can obtain the expected output value faster;

[0039] (2) Compared with conventional sliding mode variable structure control systems, the adaptive sliding mode gain coefficient designed by this method can reach the sliding mode surface within a finite time, so that the derivative of the state vector matrix of the humanoid robot joint memory motor system converges to within a finite time, which can significantly improve the system response speed and greatly reduce the overshoot, effectively improve the speed information extraction and control effect of the memory motor, and can effectively enter the sliding mode and reduce the motor system jitter. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0042] In order to improve the estimation accuracy of speed sensorless control of humanoid robot joint memory motor under high dynamic conditions, the present invention proposes a speed sensorless control method for humanoid robot joint memory motor based on second-order generalized integral sliding mode. Figure 1 As shown, the details are as follows:

[0043] Firstly, a control method is given by observing the stator current component i of the d-axis of the humanoid robot joint memory motor. d and the q-axis stator current component i q , design speed, electromagnetic torque, flux linkage and d-axis stator voltage component u d and the q-axis stator voltage component u q The dynamic mathematical model of the current is used to adjust the electrical angular velocity of the humanoid robot joint memory motor through the traditional PI controller, and the d-axis stator current component i is obtained. d and the q-axis stator current component i q The control model is:

[0044]

[0045] Among them, ω c is the electrical angular velocity, t is the time, R is the resistance, L is the inductance, is u d The estimated value of is u q The estimated value of is i dThe estimated value of is i q The estimated value of .

[0046] According to the equivalent control principle, the electrical angular velocity ω is obtained c Observed value

[0047]

[0048] Among them, K p and K i are the two parameters involved in PI control, ψ f is the magnetic flux, given by the electrical angular velocity ω c Observed value Define the sliding surface s;

[0049]

[0050] Through the second-order generalized integral sliding mode control strategy, the controller structure is designed as follows:

[0051]

[0052] Among them, sign(·) is the sign function, u is the control rate, is the derivative of u, is the derivative of s; k v is the adaptive sliding mode gain coefficient;

[0053] In the process of speed regulation of humanoid robot joint memory motor, it is necessary to add a suitable adaptive sliding mode gain coefficient k v Ensure the accuracy of the control memory motor system so that it has good performance at different speeds, as shown in formula (5),

[0054]

[0055] Among them, k min is the minimum sliding mode gain, k max is the highest sliding mode gain, ensuring that the memory motor still has good robustness when the humanoid robot joints are frequently running; l is the adaptive gain adjustment parameter, is i α The estimated value of α is the α-axis current, is i β The estimated value of β is the β-axis current;

[0056] According to the spatial structure of the sliding surface, the second-order sliding mode switching function reduces the convergence range of the memory motor system and limits the boundary of the memory motor system's motion, thereby achieving the purpose of suppressing chattering.

[0057] When the humanoid robot is operating normally, the mechanical time constant of the memory motor system is much larger than its electromagnetic time constant. Therefore, within a sampling period, it is necessary to select the state vector matrix η and complete the stability analysis according to the Lyapunov theorem.

[0058]

[0059] Where t is time and superscript T is transposition. If the designed adaptive sliding mode gain coefficient k v When the sliding surface s is reached in a finite time, the derivative of the state vector matrix of the humanoid robot joint memory motor system is Converges to 0 in finite time.

[0060]

[0061] Where P is the state matrix of the memory motor system;

[0062] Select the electrical angular velocity ω c Observed value And deriving formula (7) we get:

[0063]

[0064] Among them, η T is the transposed matrix of η, yes The transposed matrix of

[0065] By determining the stability of η, when η is a positive definite matrix, the matrix is a negative definite matrix of the state vector matrix η, then The positive definite condition is satisfied, so the memory motor system is asymptotically stable at the equilibrium point, that is, the designed second-order generalized integral sliding mode control strategy is stable, which means that the estimated electrical angular velocity observation value Make the humanoid robot joint memory motor run smoothly.

[0066] The present invention proposes a second-order generalized integral sliding mode speed sensorless control method. First, by observing the d-axis stator current component and the q-axis stator current component of the joint memory motor of the humanoid robot, the speed, electromagnetic torque, magnetic flux and the d-axis stator voltage component and the q-axis stator voltage component are designed, and the relationship between the current and the electrical angular velocity of the joint memory motor is established. Then, according to the equivalent control principle, the observed value of the electrical angular velocity is obtained, and the sliding surface is defined by the observed value of the electrical angular velocity. Through the second-order generalized integral sliding mode control strategy, the controller structure is designed, and a suitable adaptive sliding mode gain coefficient is added to ensure the accuracy of the control system, so that good performance can be achieved at different speeds. Finally, according to the Lyapunov theorem, the stability analysis is completed, and it is proved that the estimated observed value of the electrical angular velocity makes the joint memory motor of the humanoid robot run smoothly.

[0067] The above description is only a specific implementation of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a technician familiar with the technical field within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention.

Claims

1. A speed sensorless control method for a humanoid robot joint memory motor, characterized in that: Firstly, by observing the d-axis stator current component and the q-axis stator current component of the humanoid robot joint memory motor, the speed, electromagnetic torque, flux linkage, d-axis stator voltage component and q-axis stator voltage component are designed, and the relationship between the current and electrical angular velocity of the joint memory motor is established. Then, the observed value of the electrical angular velocity is obtained according to the equivalent control principle, and the sliding surface is defined by the observed value of the electrical angular velocity. The controller structure is designed through the second-order generalized integral sliding mode control strategy, and an adaptive sliding mode gain coefficient is added. Finally, the stability analysis was completed according to the Lyapunov theorem, and the estimated electrical angular velocity observation value made the joint memory motor of the humanoid robot run smoothly; The observed value of the electrical angular velocity is obtained according to the equivalent control principle, and the sliding surface is defined by the observed value of the electrical angular velocity. The controller structure is designed through the second-order generalized integral sliding mode control strategy, and an adaptive sliding mode gain coefficient is added; the details are as follows: According to the equivalent control principle, the electrical angular velocity ω is obtained c Observed value Among them, i d 、i q are the d-axis stator current component and q-axis stator current component of the humanoid robot joint memory motor, is i d The estimated value of is i q The estimated value of K p and K i are the two parameters involved in PI control, ψ f is the magnetic flux, given by the electrical angular velocity ω c Observed value Define the sliding surface s, L is the inductance; Through the second-order generalized integral sliding mode control strategy, the controller structure is designed as follows: Among them, sign(·) is the sign function, u is the control rate, is the derivative of u, is the derivative of s; k v is the adaptive sliding mode gain coefficient; In the process of speed regulation of the joint memory motor of the humanoid robot, an adaptive sliding mode gain coefficient k is added. v To ensure the accuracy of the control memory motor system, as shown in formula (4), Among them, k min is the minimum sliding mode gain, k max is the highest sliding mode gain, l is the adaptive gain adjustment parameter, is i α The estimated value of α is the α-axis current, is i β The estimated value of β is the β-axis current; According to the spatial structure of the sliding surface, the second-order sliding mode switching function limits the boundary of the memory motor system's motion by narrowing the convergence range of the memory motor system.

2. The speed sensorless control method of a humanoid robot joint memory motor according to claim 1, characterized in that: The relationship between the current and electrical angular velocity of the joint memory motor is established as follows: By observing the stator current component i of the d-axis of the humanoid robot joint memory motor d and the q-axis stator current component i q , design speed, electromagnetic torque, flux linkage and d-axis stator voltage component u d and the q-axis stator voltage component u q The dynamic mathematical model of the current is used to adjust the electrical angular velocity of the humanoid robot joint memory motor through the PI controller, and the d-axis stator current component i is obtained. d and the q-axis stator current component i q The control model is: Among them, ω c is the electrical angular velocity, t is the time, R is the resistance, is u d The estimated value of is u q The estimated value of .

3. The speed sensorless control method of a humanoid robot joint memory motor according to claim 2, characterized in that: The stability analysis is completed according to the Lyapunov theorem, and the estimated electrical angular velocity observation value makes the joint memory motor of the humanoid robot run smoothly; the details are as follows: In a sampling period, the state vector matrix η is selected and the stability analysis is completed according to the Lyapunov theorem. Where t is time and superscript T is transpose; If the designed adaptive sliding mode gain coefficient k v When the sliding surface s is reached in a finite time, the derivative of the state vector matrix of the humanoid robot joint memory motor system is Converges to 0 in a finite time; Where P is the state matrix of the memory motor system; Select the electrical angular velocity ω c Observed value And then derive formula (7) to obtain: Among them, η T is the transposed matrix of η, yes The transposed matrix of The stability of η is determined. When η is a positive definite matrix, the matrix is a negative definite matrix of the state vector matrix η, then The positive definite condition is satisfied, and the memory motor system is stable at the equilibrium point, that is, the designed second-order generalized integral sliding mode control strategy is stable, which means that the estimated electrical angular velocity observation value Make the humanoid robot joint memory motor run smoothly.

Images