Robot joint module position sensorless torque disturbance suppression control method

Abstract

The application discloses a kind of robot joint module position sensorless torque disturbance suppression control methods, comprising: obtaining permanent magnet synchronous motor current and voltage, coordinate transformation is carried out;In synchronous rotating coordinate system, establish permanent magnet synchronous motor d-axis model, extract speed related state quantity;Mechanical subsystem model is established by introducing torque disturbance;The differential equation of unified state is constructed, and the acquisition of position, speed and torque disturbance is unified into the same state space;Unified observer based on sliding mode theory is constructed, d-axis current, speed and extended disturbance state are updated synchronously, and the estimated value of electric angular velocity and the estimated value of electric angular position are output simultaneously to obtain comprehensive disturbance observation value;Torque look-ahead with forgetting factor is constructed, disturbance prediction value is calculated and speed variation is estimated;PI compensation regulator with zero speed fluctuation as target is constructed, according to the estimated speed variation, output torque current compensation, superimposed into given q-axis current, form final control quantity.

Description

Technical Field

[0001] This invention relates to the field of permanent magnet motor control technology, and more specifically to a sensorless torque disturbance suppression control method for robot joint modules. Background Technology

[0002] Sensorless control, as an advanced motor control strategy, has broad application prospects in high-performance servo drive scenarios such as robot joint modules. For robot joint modules, the drive system not only requires the motor to achieve low-speed operation, but also requires good stability, disturbance rejection and control accuracy under extremely low-speed conditions. Therefore, the research on sensorless control methods in the low-speed range and its ability to improve torque disturbance rejection has always been an important direction in this field.

[0003] Existing sensorless control methods for permanent magnet synchronous motors can generally be divided into two categories based on their applicable speed range: open-loop IF starting method or high-frequency signal injection method in the zero-low speed range, and observer method based on back EMF or flux linkage estimation in the medium-high speed range.

[0004] The IF starting method is essentially an open-loop control, making it difficult to balance control accuracy and operational efficiency. It typically fails to meet the requirements of robot joint modules for dynamic response and stability at extremely low speeds. While the high-frequency signal injection method is suitable for position identification at low speeds, it requires injecting additional high-frequency excitation signals into the motor, which not only introduces electromagnetic noise but also reduces energy utilization efficiency. Furthermore, this method usually relies on the salient polarity of the motor, limiting its applicability. In addition, methods based on back EMF and flux linkage estimation typically require intermediate signals to indirectly extract rotational speed and position information. Although they perform well in the medium-to-high speed range, the weak back EMF amplitude and low signal-to-noise ratio at extremely low speeds lead to decreased observation accuracy, making it difficult to achieve reliable position estimation.

[0005] In addition, in the application scenarios of robot joint modules, especially direct-drive joint modules, due to the lack of buffering and filtering effect of mechanical transmission links, the speed fluctuations generated by the motor when running at extremely low speeds will directly affect the load end, causing noise and vibration, and in severe cases, may even cause load damage.

[0006] Existing solutions for suppressing cogging torque and torque ripple can be broadly categorized into two types: reducing cogging torque by optimizing the motor body structure, and mitigating the impact of disturbances on system performance by improving control strategies. The former is often limited by cost, manufacturing process, and structural implementation conditions, making it difficult to be widely applied to existing robot joint module platforms; the latter is more flexible and has better engineering application value.

[0007] However, most existing control strategies design sensorless control and disturbance suppression separately, usually using a cascaded observer structure. That is, speed or position observation is completed first, and then torque or disturbance observer is built based on the results. This type of method is prone to introducing additional phase lag, especially under extremely low speed conditions, where the viscous damping term in the mechanical equation is small. Traditional cascaded disturbance observation structures often suffer from inaccurate estimation and slow dynamic response, which in turn affects the disturbance compensation effect.

[0008] Therefore, how to achieve reliable estimation of rotor speed and position without relying on mechanical position sensors and without injecting high-frequency signals, and to uniformly observe and compensate for equivalent torque disturbances, so as to improve the stability, anti-disturbance performance and control accuracy of robot joint modules under extremely low-speed conditions, is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0009] In view of the above problems, the present invention provides a sensorless torque disturbance suppression control method for robot joint modules that overcomes or at least partially solves the above problems.

[0010] To achieve the above objectives, the present invention adopts the following technical solution:

[0011] A sensorless torque disturbance suppression control method for robot joint modules includes: S1. Obtain the three-phase current and d-axis voltage of the permanent magnet synchronous motor; S2. Based on the rotor's electrical angular position, perform coordinate transformation to convert the three-phase current into d-axis and q-axis currents; S3. Select the d-axis current equation of the permanent magnet synchronous motor as the basis for the observer modeling. In the synchronous rotating coordinate system, establish the d-axis model of the permanent magnet synchronous motor and extract the speed-related state variables. S4. Introduce torque disturbance and further introduce mechanical motion equations to establish a mechanical subsystem model; S5. Define the state variables of the electrical subsystem and the mechanical subsystem respectively, and introduce the unknown disturbance as an extended state variable into the unified state space model to construct the differential equation of the unified state, and unify the acquisition of position, velocity and torque disturbances into the same state space framework; S6. Based on the differential equation of the unified state of the system, a fusion unified observer based on sliding mode theory is constructed, using the d-axis current estimation error as the only global driving signal to synchronously update the d-axis current, rotational speed and extended disturbance state. S7. Stability design of the fusion unified observer is carried out by analyzing and calculating the sliding mode gain tuning rule using the Lyapunov stability method. S8. Using a fusion-type unified observer, output the estimated electric angular velocity and electric angular position values, and simultaneously output the extended state estimate to obtain the comprehensive disturbance observation value; S9. Construct a torque look-ahead device with a forgetting factor based on the comprehensive disturbance observations, calculate the disturbance prediction value and estimate the speed change; S10. Construct a PI compensation regulator with zero speed fluctuation as the target, and output torque current compensation amount based on the estimated speed change; S11. Feedback the estimated electrical angular position value to step S2 for coordinate transformation and sensorless closed-loop control. Add the torque current compensation to the given q-axis current to form the final control quantity to achieve low-speed and stable operation of the robot joint module.

[0012] Preferably, step S2, the method for performing coordinate transformation, is as follows:

[0013] in, i d for d shaft current, i q for q shaft current, i e For rotor position, i a , i b , i c It is a three-phase current.

[0014] Preferably, in step S3, the d-axis model of the permanent magnet synchronous motor is:

[0015] in, R s For stator resistance, L d , L q They are motors d , q Shaft inductor, oh e Electric angular velocity, i d , i q They are respectively d , q shaft current, u d This is the d-axis voltage vector.

[0016] Preferably, in step S4, the mechanical subsystem model is as follows:

[0017] in, n p For extreme logarithms, J For rotational inertia, B The viscous damping coefficient is... T e For electromagnetic torque, T L The equivalent load torque includes the external load and the equivalent disturbance referred to the mechanical side.

[0018] Preferably, in step S5, the differential equation for the unified state is specifically as follows:

[0019] in, b 01 and b 02 For characteristic gain, f 01 , f 02 These are the known disturbance terms for the electrical subsystem and the mechanical subsystem, respectively. f 1 represents the unknown disturbance term in the mechanical subsystem. u d The voltage vector along the d-axis. T e It is electromagnetic torque.

[0020] Preferably, in step S6, the fusion-type unified observer specifically refers to:

[0021] in, , , These are the estimated values ​​of the d-axis current, the estimated value of the rotor electric angular velocity, and the observed value of the combined disturbance, respectively. K sm1 , K sm2 and K sm3 is the sliding mode gain, and sgn(·) is the sign function.

[0022] Preferably, in step S7, the sliding mode gain tuning rule is as follows:

[0023] in, H eq This is the equivalent proportionality coefficient. oh 0 represents the observer bandwidth. J The moment of inertia of the motor. Bis the viscous damping coefficient.

[0024] Preferably, in step S9, the predicted disturbance value for the next time step is:

[0025] in, l Let be the forgetting factor, and 0 < λ < 1. The perturbation observation value at the current sampling time. This represents the perturbation observation value from the previous sampling time. Based on the mechanical motion equations, the change in rotational speed caused by the disturbance is estimated as follows:

[0026] in, T s To control the cycle.

[0027] Preferably, in step S10, the output of the PI compensation regulator is:

[0028] in, K p This is the proportionality coefficient. K i The integral coefficient is... This is for speed fluctuation error;

[0029] in, This is used to estimate the change in rotational speed caused by the disturbance.

[0030] Preferably, in step S11, the final control quantity is:

[0031] in, i q,ref For the original q Shaft current reference value, This refers to the torque current compensation amount output by the PI compensation regulator. i q After compensation q Shaft current given.

[0032] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a sensorless torque disturbance suppression control method for robot joint modules, which eliminates the dependence of traditional low-speed positionless control on high-frequency signal injection, and breaks the paradigm of conventional cascaded observers. It solves the phase lag problem in traditional cascaded systems by using fused observation. The present invention estimates the electric angular position and speed of the motor rotor without relying on mechanical position sensors, and simultaneously realizes the synchronous observation and compensation of multiple types of torque disturbances, thereby suppressing speed fluctuations and crawling phenomena during extremely low-speed operation, and effectively improving the low-speed stability, anti-disturbance performance and control accuracy of robot joint modules. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0034] Figure 1 This is a schematic diagram of a sensorless torque disturbance suppression control method for robot joint modules provided in an embodiment of the present invention. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] This invention discloses a sensorless torque disturbance suppression control method for robot joint modules, such as... Figure 1 ,include: S1. Obtain the three-phase current and d-axis voltage of the permanent magnet synchronous motor; the three-phase current of the permanent magnet synchronous motor can generally be obtained by direct sampling; the d-axis voltage can be obtained by coordinate transformation of the three-phase voltage; S2. Based on the rotor's electrical angular position, perform coordinate transformation to convert the three-phase current into d-axis and q-axis currents; S3. Select the d-axis current equation of the permanent magnet synchronous motor as the basis for the observer modeling. In the synchronous rotating coordinate system, establish the d-axis model of the permanent magnet synchronous motor and extract the speed-related state variables. S4. Introduce torque disturbance and further introduce mechanical motion equations to establish a mechanical subsystem model; S5. Define the state variables of the electrical subsystem and the mechanical subsystem respectively, and introduce the unknown disturbance as an extended state variable into the unified state space model to construct the differential equation of the unified state, and unify the acquisition of position, velocity and torque disturbances into the same state space framework; S6. Based on the differential equation of the unified state of the system, a fusion unified observer based on sliding mode theory is constructed, using the d-axis current estimation error as the only global driving signal to synchronously update the d-axis current, rotational speed and extended disturbance state. S7. Stability design of the fusion unified observer is carried out by analyzing and calculating the sliding mode gain tuning rule using the Lyapunov stability method. S8. Using a fusion-type unified observer, output the estimated electric angular velocity and electric angular position values, and simultaneously output the extended state estimate to obtain the comprehensive disturbance observation value; S9. Construct a torque look-ahead device with a forgetting factor based on the comprehensive disturbance observations, calculate the disturbance prediction value and estimate the speed change; S10. Construct a PI compensation regulator with zero speed fluctuation as the target, and output torque current compensation amount based on the estimated speed change; S11. Feedback the estimated electrical angular position value to step S2 for coordinate transformation and sensorless closed-loop control. Add the torque current compensation to the given q-axis current (the q-axis current control quantity output by the original speed loop PI) to form the final control quantity to achieve low-speed and stable operation of the robot joint module.

[0037] In practical applications, the electrical parameters of a permanent magnet motor can be selected as follows: inductance is... L d =L q =0.3mH, resistance is R s =0.1Ω, rotor flux linkage is ψ f = 0.0875 Wb ， The number of permanent magnets is n.p. =3, rated speed is 300 rad / s, control cycle is T s = 0.0002 s, moment of inertia of the motor J =0.031, viscous damping coefficient B =0.008.

[0038] To further implement the above technical solution, step S2, the coordinate transformation includes performing Clarke transformation and Park transformation sequentially, specifically as follows:

[0039] in,i d for d shaft current, i q for q shaft current, i e The rotor's electrical angle position, i a , i b , i c It is a three-phase current.

[0040] To further implement the above technical solution, in step S3, the d-axis model of the permanent magnet synchronous motor is as follows:

[0041] in, R s For stator resistance, L d , L q They are motors d , q Shaft inductor, oh e Electric angular velocity, i d , i q They are respectively d , q shaft current, u d This is the d-axis voltage vector.

[0042] In this embodiment, the d-axis current model contains the rotational speed information to be observed. Therefore, instead of using a high-frequency signal injection method to obtain position information, the rotational speed-related state variables are directly extracted from the mathematical model of the permanent magnet synchronous motor.

[0043] To further implement the above technical solution, in step S4, in order to synchronously observe the torque disturbance during motor operation, a mechanical subsystem model is introduced as follows:

[0044] in, n p For extreme logarithms, J For rotational inertia, B The viscous damping coefficient is... T e For electromagnetic torque, T L For the equivalent load torque, in scenarios where the robot joint module operates at extremely low speeds, T LIt includes not only the effects of external loads, but also the equivalent disturbances referred to the mechanical side, such as cogging torque, inverter nonlinearity error, and parameter perturbations, all of which are treated as disturbances to be observed.

[0045] To further implement the above technical solution, in step S5, in order to simultaneously achieve sensorless control and disturbance observation within a single observer framework, the state variables of the electrical subsystem and the mechanical subsystem are defined respectively. For the electrical subsystem, u d Treat it as system input, and set 1 / L d Treat it as a characteristic gain and the velocity coupling term oh e L q i q Treat it as a known disturbance; for the mechanical subsystem, T e Treat it as system input, n p / J Consider it as characteristic gain, and - Wow e / J Treat it as a known disturbance, and T L Treat it as an unknown disturbance; The specific system state equation is expressed as follows:

[0046] in, b 01 and b 02 Indicates characteristic gain, f 01 and f 02 These represent the known disturbance terms of the two systems mentioned above. u 1 and u 2 represents the system input of the electrical subsystem and the mechanical subsystem, respectively. f 1 represents an unknown disturbance term in the mechanical subsystem; Unknown disturbances are introduced as extended state variables to form a unified state-space model, and the state variables are further defined. x 1. x 2 and x 3 is:

[0047] Therefore, the differential equation for the unified state is as follows:

[0048] in, b 01 and b 02 For characteristic gain, f 01 , f 02 These are the known disturbance terms for the electrical subsystem and the mechanical subsystem, respectively. f 1 represents the unknown disturbance term in the mechanical subsystem. u d The voltage vector along the d-axis. T e It is electromagnetic torque.

[0049] To further implement the above technical solution, in step S6, the SM-UDSIP observer based on sliding mode theory is specifically as follows:

[0050] in, , , These are the estimated values ​​of the d-axis current, the estimated value of the rotor electric angular velocity, and the observed value of the combined disturbance, respectively. K sm1 , K sm2 and K sm3 is the sliding mode gain, and sgn(·) is the sign function.

[0051] The fusion-based unified observer SM-UDSIP in this embodiment differs from traditional cascaded observers. It no longer uses the method of "first estimating velocity and position, and then constructing disturbance observers in series," but instead acquires data synchronously through the fusion-based unified observer. and The estimated value of the rotor's electric angular velocity and the comprehensive disturbance observation value solve the phase lag problem in the traditional cascade structure.

[0052] To further implement the above technical solution, the sliding mode gain tuning rule in step S7 is as follows:

[0053] in, H eq This is the equivalent proportionality coefficient. oh 0 represents the observer bandwidth.

[0054] In this embodiment, the specific method for stability design of the fusion unified observer is as follows: To ensure the convergence of the constructed observer, the Lyapunov stability method is used for analysis, and the Lyapunov function is constructed as follows:

[0055] in, The sliding surface constructed to account for current estimation errors. , , used to represent the observation error of each state quantity; Differentiating the above equation, we get:

[0056] According to the sliding mode arrival condition, to ensure stable convergence of the observer, the following must be satisfied:

[0057] Substituting the expression for the fusion unified observer SM-UDSIP, we get:

[0058] Furthermore, we can obtain:

[0059] According to Lyapunov stability theory, for V, its value is always greater than 0; while for dV / dt, the first term is always less than or equal to 0. To ensure that the stability condition always holds, the following sufficient condition must be satisfied:

[0060] When the system converges to the sliding surface and maintains stable sliding motion, we have Substitute From the expression, we can obtain:

[0061] Among them, sgn[(·)] eq It represents the equivalent switching function, serving as a carrier for cross-domain electromechanical information transmission and providing driving force for the synchronous update of rotor speed and extended state; To facilitate stability analysis of the reduced-order system, an equivalent scaling factor is defined. H eq for:

[0062] Substitute the equivalent control item x 2 and x From the estimated error dynamic equation of 3, the state-space equation of the reduced-order error system can be obtained as follows:

[0063] Its characteristic equation is:

[0064] According to the Routh–Hurwitz stability criterion, to ensure the global asymptotic stability of sliding mode motion, all poles must lie strictly within the left half of the complex plane, and all coefficients must be strictly positive; because B / J >0 always holds true, therefore it is necessary to ensure K sm2 H eq and K sm3 H eq All are greater than zero; using the linear pole placement method, the desired poles in the above equation can be uniformly placed at - oh 0 locations oh 0 represents the observer bandwidth; from this, it is deduced that... K sm2 and K sm3 The tuning rules.

[0065] In this embodiment, in step S8, the fusion-type unified observer constructed in step S6 can directly output the estimated electric angular velocity value:

[0066] Then, the estimated electric angular position is obtained by integration:

[0067] This means converges to That is, the obtained electrical angular position estimate is fed back to step S2 for coordinate transformation, converting the three-phase current into d-axis and q-axis currents; This embodiment uses the method of "directly extracting the rotation speed and integrating to obtain the position" to complete sensorless control; The extended state estimate, i.e., the combined disturbance observation, is output by the fusion unified observer:

[0068] Based on the modeling methods in steps S5 and S6, The estimation results of the comprehensive disturbance torque characterizing the robot joint module during extremely low-speed operation include at least external load disturbance, cogging torque disturbance, equivalent disturbance caused by inverter nonlinearity, and the influence of some parameter mismatches. Therefore, this embodiment does not connect a torque observer separately after the speed estimation is completed, but obtains the speed estimation and disturbance torque estimation simultaneously within a single observer framework, thereby improving the time delay problem of traditional cascaded observation schemes.

[0069] To further implement the above technical solution, in step S9, the predicted disturbance value for the next moment is:

[0070] in, l Let be the forgetting factor, and 0 < λ < 1. The perturbation observation value at the current sampling time. This represents the perturbation observation value from the previous sampling time. Based on the mechanical motion equations, the change in rotational speed caused by the disturbance is estimated as follows:

[0071] in, T s To control the cycle.

[0072] To further implement the above technical solution, in step S10, the output of the PI compensation regulator is:

[0073] in, K p This is the proportionality coefficient. K i The integral coefficient is... This is for speed fluctuation error;

[0074] in, Estimate the change in rotational speed caused by the disturbance; The output of the PI compensation regulator is applied to the torque current control command, thereby achieving error-free suppression of predicted speed fluctuations.

[0075] To further implement the above technical solution, in step S11, the final control quantity is:

[0076] in, i q,ref For the original q The shaft current reference value is the output of the original velocity loop PI of the vector control dual closed loop. This refers to the torque current compensation amount output by the PI compensation regulator. i q After compensation q Shaft current given.

[0077] This embodiment enables the motor to operate smoothly at low speeds without relying on mechanical position sensors, without interrupting control, and without requiring high-frequency signal injection, and improves the ability to suppress torque disturbances.

[0078] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0079] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A sensorless torque disturbance suppression control method for robot joint modules, characterized in that, include: S1. Obtain the three-phase current and d-axis voltage of the permanent magnet synchronous motor; S2. Based on the rotor's electrical angular position, perform coordinate transformation to convert the three-phase current into d-axis and q-axis currents; S3. Select the d-axis current equation of the permanent magnet synchronous motor as the basis for the observer modeling. In the synchronous rotating coordinate system, establish the d-axis model of the permanent magnet synchronous motor and extract the speed-related state variables. S4. Introduce torque disturbance and further introduce mechanical motion equations to establish a mechanical subsystem model; S5. Define the state variables of the electrical subsystem and the mechanical subsystem respectively, and introduce the unknown disturbances as extended state variables into the unified state space model to construct the differential equations of the unified state, and unify the acquisition of position, velocity and torque disturbances into the same state space framework; S6. Based on the differential equation of the unified state of the system, a fusion unified observer based on sliding mode theory is constructed, using the d-axis current estimation error as the only global driving signal to synchronously update the d-axis current, rotational speed and extended disturbance state. S7. Stability design of the fusion unified observer is carried out by analyzing and calculating the sliding mode gain tuning rule using the Lyapunov stability method. S8. Using a fusion-type unified observer, output the estimated electric angular velocity and electric angular position values, and simultaneously output the extended state estimate to obtain the comprehensive disturbance observation value; S9. Construct a torque look-ahead device with a forgetting factor based on the comprehensive disturbance observations, calculate the disturbance prediction value and estimate the speed change; S10. Construct a PI compensation regulator with zero speed fluctuation as the target, and output torque current compensation amount based on the estimated speed change; S11. Feedback the estimated electrical angular position value to step S2 for coordinate transformation and sensorless closed-loop control. Add the torque current compensation to the given q-axis current to form the final control quantity to achieve low-speed and stable operation of the robot joint module.

2. The sensorless torque disturbance suppression control method for robot joint modules as described in claim 1, characterized in that, Step S2, the method for performing coordinate transformation is as follows: in, i d for d shaft current, i q for q shaft current, θ e For rotor position, i a , i b , i c It is a three-phase current.

3. The sensorless torque disturbance suppression control method for robot joint modules as described in claim 1, characterized in that, In step S3, the d-axis model of the permanent magnet synchronous motor is as follows: in, R s For stator resistance, L d , L q They are motors d , q Shaft inductor, ω e Electric angular velocity, i d , i q They are respectively d , q shaft current, u d This is the d-axis voltage vector.

4. The sensorless torque disturbance suppression control method for robot joint modules as described in claim 1, characterized in that, In step S4, the mechanical subsystem model is as follows: in, n p For extreme logarithms, J For rotational inertia, B The viscous damping coefficient is... T e For electromagnetic torque, T L The equivalent load torque includes the external load and the equivalent disturbance referred to the mechanical side.

5. The sensorless torque disturbance suppression control method for robot joint modules as described in claim 1, characterized in that, In step S5, the differential equation for the unified state is as follows: in, b 01 and b 02 For characteristic gain, f 01 , f 02 These are the known disturbance terms for the electrical subsystem and the mechanical subsystem, respectively. f 1 represents the unknown disturbance term in the mechanical subsystem. u d The voltage vector along the d-axis. T e It is electromagnetic torque.

6. The sensorless torque disturbance suppression control method for robot joint modules as described in claim 5, characterized in that, In step S6, the fusion-type unified observer is specifically as follows: in, , , These are the estimated values ​​of the d-axis current, the estimated value of the rotor electric angular velocity, and the observed value of the combined disturbance, respectively. K sm1 , K sm2 and K sm3 is the sliding mode gain, and sgn(·) is the sign function.

7. The sensorless torque disturbance suppression control method for robot joint modules as described in claim 1, characterized in that, In step S7, the sliding mode gain tuning rule is as follows: in, H eq This is the equivalent proportionality coefficient. ω 0 represents the observer bandwidth. J The moment of inertia of the motor. B is the viscous damping coefficient.

8. The sensorless torque disturbance suppression control method for robot joint modules as described in claim 1, characterized in that, In step S9, the predicted disturbance value for the next time step is: in, λ Let be the forgetting factor, and 0 < λ < 1. The perturbation observation value at the current sampling time. This represents the perturbation observation value from the previous sampling time. Based on the mechanical motion equations, the change in rotational speed caused by the disturbance is estimated as follows: in, T s To control the cycle.

9. The sensorless torque disturbance suppression control method for robot joint modules as described in claim 1, characterized in that, In step S10, the output of the PI compensation regulator is: in, K p This is the proportionality coefficient. K i The integral coefficient is... This is for speed fluctuation error; in, This is used to estimate the change in rotational speed caused by the disturbance.

10. The sensorless torque disturbance suppression control method for robot joint modules as described in claim 1, characterized in that, In step S11, the final control variable is: in, i q,ref For the original q Shaft current reference value, This refers to the torque current compensation amount output by the PI compensation regulator. i q After compensation q Shaft current given.

Images