Robot position control methods, products and equipment based on motor current prediction

By observing the disturbance of the permanent magnet synchronous motor in real time through a linear cascade disturbance observer and correcting the state variable prediction, combined with deadbeat control and delay compensation, the problem of decreased position control accuracy of humanoid robots in the pumping station environment was solved, and high-precision and robust position control was achieved.

CN123074366APending Publication Date: 2026-11-10NORTHWEST ENGINEERING CORPORATION LIMITED
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611379458.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-07
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

In pumping station environments, the parameters of permanent magnet synchronous motors of humanoid robots are susceptible to moisture and electromagnetic interference, leading to a decrease in position control accuracy. Furthermore, existing model predictive control methods rely on fixed parameters and cannot adapt to high-frequency start-stop and load changes, resulting in lag in dynamic response and limb tremors.

Method used

A linear cascade disturbance observer is used to observe the disturbance of the permanent magnet synchronous motor in real time. The state variable prediction is corrected by the disturbance observation value, and the reference voltage is determined to reduce the impact of parameter mismatch. Combined with deadbeat control and delay compensation, the position control accuracy is improved.

Benefits of technology

Maintaining high-precision position control while parameters change dynamically improves the robot's robustness and dynamic performance, enabling it to adapt to complex working conditions and meet the requirements for high reliability and high-precision motion.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123074366A_ABST
    Figure CN123074366A_ABST
Patent Text Reader

Abstract

This disclosure provides a robot position control method, product, and device based on motor current prediction, relating to the field of motor control technology. The method includes: acquiring the state variables and control signals of the permanent magnet synchronous motor (PMSM) at the current moment during continuous position control of the robot by a PMSM; observing the total disturbance currently experienced by the PMSM based on a preset linear cascade disturbance observer to obtain the disturbance observation value at the current moment; calculating the predicted state variables of the PMSM at the next moment based on the disturbance observation value, the state variables at the current moment, and the voltage; and determining the reference voltage of the PMSM based on the predicted state variables at the next moment and a preset target current, so that the PMSM can perform position control of the robot at the next moment based on the reference voltage. This disclosure improves the position control accuracy of the robot.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This disclosure relates to the field of motor control technology, and more specifically, to a robot position control method, product, and device based on motor current prediction. Background Technology

[0002] With the advancement of smart water management, humanoid robots are being gradually applied to complex tasks such as pump station inspection and emergency intervention. The joint drives of humanoid robots mostly use permanent magnet synchronous motors (PMSMs), which require extremely high precision in position control.

[0003] Currently, position control widely employs Model Predictive Control (MPC), a method that heavily relies on accurate motor parameters to construct predictive models. However, pumping station environments are often humid, confined, and subject to strong electromagnetic interference. Humanoid robots experience frequent start-stop cycles and sudden load changes, making motor parameters susceptible to fluctuations. Parameter mismatch further degrades the accuracy of position control. Summary of the Invention

[0004] This disclosure provides a robot position control method, product, and device based on motor current prediction, which at least partially solves the technical problem of low robot position control accuracy.

[0005] According to a first aspect of this disclosure, a robot position control method based on motor current prediction is provided, comprising: During the continuous position control of the robot by the permanent magnet synchronous motor, the state variables and control signals of the permanent magnet synchronous motor at the current moment are acquired. Based on a preset linear cascade disturbance observer, the total disturbance currently experienced by the permanent magnet synchronous motor is observed to obtain the disturbance observation value at the current moment. Based on the disturbance observation, the current state variables, and the control signal, calculate the predicted state variables of the permanent magnet synchronous motor at the next moment; Based on the predicted state variables and the preset target current at the next moment, the reference voltage of the permanent magnet synchronous motor at the next moment is determined, so that the permanent magnet synchronous motor can perform position control on the robot based on the reference voltage at the next moment.

[0006] According to a second aspect of this disclosure, a robot position control device based on motor current prediction is provided, comprising: The information acquisition module is used to acquire the state variables and control signals of the permanent magnet synchronous motor at the current moment during the continuous position control of the robot by the permanent magnet synchronous motor. The disturbance observation module is used to observe the total disturbance currently experienced by the permanent magnet synchronous motor based on a preset linear cascade disturbance observer, and obtain the disturbance observation value at the current moment. The current prediction module is used to calculate the predicted state variables of the permanent magnet synchronous motor at the next moment based on the disturbance observation value, the state variables at the current moment, and the control signal. The position control module is used to determine the reference voltage of the permanent magnet synchronous motor at the next moment based on the predicted current and the preset target current at the next moment, so that the permanent magnet synchronous motor can perform position control on the robot based on the reference voltage at the next moment.

[0007] According to a third aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method of the first aspect described above and possible implementations thereof.

[0008] According to a fourth aspect of this disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform the method of the first aspect and possible implementations thereof by executing the executable instructions.

[0009] The technical solution disclosed herein has the following beneficial effects: This disclosure utilizes a linear cascaded disturbance observer to monitor the disturbances currently received by the permanent magnet synchronous motor in real time, and incorporates the disturbance observation values ​​into the prediction of the state variables at the next moment. Based on this, the reference voltage of the permanent magnet synchronous motor at the next moment is determined, realizing the correction of the prediction results according to the actual disturbance. By introducing disturbance observation values, the prediction of the state variables and the reference voltage at the next moment no longer depend solely on fixed, known parameters, reducing the impact of parameter mismatch on position control accuracy. This allows the permanent magnet synchronous motor servo system to maintain high-precision position control even when parameters change dynamically, better meeting the fine motion requirements of humanoid robots.

[0010] This disclosure solves the problems in related technologies, such as the inability to perceive and compensate for disturbances in real time, resulting in lag in dynamic response, speed drops, and limb tremors. It meets the high reliability and high precision motion requirements of robots in smart water pumping stations and is better able to adapt to the complex and varied operating conditions of humanoid robots in smart water pumping stations. While improving position control accuracy, it also enhances the robustness, dynamic performance, and safety of humanoid robots, especially adapting to position control scenarios using multi-motor collaborative permanent magnet synchronous motor servo systems. It addresses the shortcomings of related technologies in terms of accuracy, disturbance rejection, safety, and collaborative control in humanoid robot applications, providing core technical support for the development of high-performance humanoid robot joint servo systems. Attached Figure Description

[0011] Figure 1 A flowchart illustrating a robot position control method based on motor current prediction according to an embodiment of this disclosure is shown. Figure 2 A flowchart illustrating an embodiment of the present disclosure of obtaining a preset linear cascaded perturbation observer is shown. Figure 3 A flowchart illustrating one embodiment of this disclosure for obtaining perturbation observations is shown; Figure 4 A flowchart illustrating one embodiment of the present disclosure for obtaining a predicted current is shown; Figure 5 A flowchart illustrating one method for determining a reference voltage according to an embodiment of this disclosure is shown; Figure 6a A flowchart illustrating a constraint reference voltage in an embodiment of this disclosure is shown; Figure 6b A flowchart illustrating a simulation experiment verification of an embodiment of this disclosure is shown; Figure 6c A graph illustrating a simulation experiment verification of an embodiment of this disclosure is shown. Figure 7 A schematic diagram of a robot position control device based on motor current prediction is shown in an embodiment of this disclosure; Figure 8 A schematic diagram of an electronic device according to one embodiment of the present disclosure is shown. Detailed Implementation

[0012] Exemplary embodiments of this disclosure will be described more fully below with reference to the accompanying drawings.

[0013] The accompanying drawings are schematic illustrations of this disclosure and are not necessarily drawn to scale. Some block diagrams shown in the drawings may be functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in hardware modules or integrated circuits, or in networks, processors, or microcontrollers. Implementations can be carried out in various forms and should not be construed as limited to the examples set forth herein. The features, structures, or characteristics described in this disclosure can be combined in any suitable manner in one or more implementations. In the following description, numerous specific details are provided to give a thorough description of one embodiment of this disclosure. However, those skilled in the art will recognize that one or more specific details may be omitted when implementing the technical solutions provided in an embodiment of this disclosure, or other methods, components, apparatuses, steps, etc., may be used to replace one or more specific details.

[0014] With the deepening of smart water management construction, pumping stations, as core nodes of water supply, drainage, and water circulation systems, face increasingly urgent demands for automated and intelligent operation and maintenance. In recent years, humanoid robots, due to their flexible movement capabilities and multi-task adaptability, have been gradually explored and applied to smart water pumping stations, undertaking complex tasks such as equipment inspection, valve operation, and troubleshooting abnormal conditions. The core functions of humanoid robots, such as joint actuation and limb motion positioning, highly rely on the position control of permanent magnet synchronous motor servo systems. Their control performance directly determines the robot's motion accuracy and operational stability. Especially in the damp, confined, and electromagnetically interference-prone conditions of underground pumping stations, permanent magnet synchronous motor servo systems face stringent requirements such as rapid dynamic load changes and exacerbated time-varying parameters.

[0015] Currently, Multi-Objective Optimization (MPC) has been widely applied to the position control of permanent magnet synchronous motors (PMSMs) due to its multi-objective optimization characteristics and system constraint adaptability. This technology employs a control process involving delay compensation and reference voltage calculation, constructing an integrated predictive model based on the voltage, torque, and position equations of the PMSM to derive the reference voltage. However, MPC relies on precise motor parameters and suffers from insufficient robustness, specifically manifested in the following ways: First, during humanoid robot movement, the joint motors are constantly under high-frequency start-stop and variable load conditions. Parameters such as stator resistance, inductance, and permanent magnet flux linkage fluctuate nonlinearly with increasing temperature, and parameters such as moment of inertia and coefficient of friction also change due to variations in limb load. MPC (Multi-parameter Capability Control) relies entirely on fixed and precise motor parameters. The parameter mismatch problem, which is highly likely to occur in complex environments, will directly cause the model to deviate from the actual operating state of the motor, resulting in unavoidable prediction errors. This error will be propagated to the delay compensation stage, causing the predicted current and speed for the next moment based on ideal parameters to differ from the actual values, further amplifying the control deviation and ultimately increasing the positional positioning error, failing to meet the requirements of high-precision positioning.

[0016] Secondly, humanoid robots experience frequent external load surges (such as ground impacts during walking or sudden load increases when grasping objects) and internal coupling disturbances (such as torque interference during multi-joint coordinated movements). These disturbances alter the motor's output torque balance, causing the MPC's torque output calculations to become disconnected from actual requirements, resulting in lag in dynamic response. This manifests as speed drops and position overshoot, causing stuttering and tremors in the robot's limb movements, and even triggering oscillations in the joint servo system. Furthermore, the MPC's constraint boundary calculations also rely on precise parameters. Parameter mismatch can lead to constraint boundary shifts, potentially causing current over-limit damage to the driver or speed runaway in high-frequency, high-dynamic motion scenarios.

[0017] In summary, there is an urgent need for a robot position control method based on motor current prediction that can significantly improve the anti-disturbance performance and positioning accuracy under pump station conditions.

[0018] In view of one or more of the above-mentioned problems, this disclosure provides a robot position control method, product and device based on motor current prediction.

[0019] Figure 1 An exemplary flow of a robot position control method based on motor current prediction is shown, including the following steps: S110, during the process of the permanent magnet synchronous motor continuously controlling the position of the robot, the state variables and control signals of the permanent magnet synchronous motor at the current moment are acquired; S120, based on a preset linear cascade disturbance observer, observe the total disturbance currently experienced by the permanent magnet synchronous motor and obtain the disturbance observation value at the current moment; S130, Based on the disturbance observation, the current state variable and the control signal, calculate the predicted state variable of the permanent magnet synchronous motor at the next moment; S140, based on the predicted state variable and the preset target current at the next moment, determine the reference voltage of the permanent magnet synchronous motor at the next moment, so that the permanent magnet synchronous motor can perform position control on the robot based on the reference voltage at the next moment.

[0020] Based on the above method, this disclosure utilizes a linear cascaded disturbance observer to observe the disturbance currently received by the permanent magnet synchronous motor in real time, and incorporates the disturbance observation value into the prediction of the state variable at the next moment. Based on this, the reference voltage of the permanent magnet synchronous motor at the next moment is determined, realizing the correction of the prediction result according to the actual disturbance. By introducing the disturbance observation value, the prediction of the state variable at the next moment and the reference voltage no longer depend solely on fixed, known parameters, reducing the impact of parameter mismatch on position control accuracy. This allows the permanent magnet synchronous motor servo system to maintain high-precision position control even when parameters change dynamically, better meeting the fine motion requirements of humanoid robots.

[0021] This disclosure solves the problems in related technologies, such as the inability to perceive and compensate for disturbances in real time, resulting in lag in dynamic response, speed drops, and limb tremors. It meets the high reliability and high precision motion requirements of robots in smart water pumping stations and is better able to adapt to the complex and varied operating conditions of humanoid robots in smart water pumping stations. While improving position control accuracy, it also enhances the robustness, dynamic performance, and safety of humanoid robots, especially adapting to position control scenarios using multi-motor collaborative permanent magnet synchronous motor servo systems. It addresses the shortcomings of related technologies in terms of accuracy, disturbance rejection, safety, and collaborative control in humanoid robot applications, providing core technical support for the development of high-performance humanoid robot joint servo systems.

[0022] The following describes, in conjunction with one or more embodiments and related accompanying drawings, Figure 1 Each step in the process will be explained in detail.

[0023] refer to Figure 1 In step S110, during the process of the permanent magnet synchronous motor continuously controlling the position of the robot, the state variables and control signals of the permanent magnet synchronous motor at the current moment are acquired.

[0024] State variables refer to parameters of a permanent magnet synchronous motor that change over time, such as d-axis current, q-axis current, motor rotor electrical angular velocity (speed), and motor rotor electrical angle.

[0025] Control signals refer to the voltages output from the previous control cycle that act on the motor, such as the d-axis reference voltage and the q-axis reference voltage.

[0026] In addition, when predicting the current and voltage of the permanent magnet synchronous motor for robot position control, motor parameters such as stator resistance, stator inductance, permanent magnet flux linkage, number of pole pairs, rotor moment of inertia, coefficient of friction, and load torque are also required, as exemplified in Table 1.

[0027] Table 1 Motor Parameter Table

[0028] In one implementation, under ideal conditions, if disturbances to the permanent magnet synchronous motor are not considered and the motor parameters remain unchanged and perfectly matched, the continuous-time prediction model of the permanent magnet synchronous motor is as follows: (1) (2) (3) (4) in, This represents the d-axis current (in A). This represents the q-axis current (in A). This is the d-axis reference voltage (in V). This is the q-axis reference voltage (in V). Stator resistance (unit: Ω), Stator inductance (unit: H), The flux linkage of a permanent magnet (unit: Wb) The electric angular velocity of the motor rotor (in rad / s). The electric angle of the motor rotor (in rad). For extreme logarithms, Rotor moment of inertia (unit: ) ), The coefficient of friction (unit: ) ), Load torque (unit: The subscript 's' above indicates the stator.

[0029] Continue to refer to Figure 1 In step S120, based on a preset linear cascade disturbance observer, the total disturbance currently experienced by the permanent magnet synchronous motor is observed to obtain the disturbance observation value at the current moment.

[0030] The Linear Cascaded Disturbance Observer (LCDO) combines a disturbance observer with a cascade control structure, balancing disturbance observation accuracy and noise suppression capability. It can observe the total disturbance currently experienced by the permanent magnet synchronous motor and obtain the disturbance observation value.

[0031] Total disturbances include internal disturbances caused by changes in motor parameters and external load changes. Internal disturbances refer to internal disturbances caused by fluctuations in electrical parameters, changes in mechanical parameters, etc., such as torque interference during multi-joint coordinated motion. External load changes include ground impacts during robot walking and sudden increases in load when grasping objects.

[0032] When the permanent magnet synchronous motor is disturbed, the prediction model corresponding to equations (1)-(4) is not accurate enough. Based on this, equations (1)-(4) are modified as follows: (5) in, Representing current i d i q and rotational speed w e The total disturbances for each are as follows, and their respective expressions are as follows: (6) (7) (8) The variables in equation (5), except for the total disturbance, are all measurable state variables or motor parameters. Therefore, as long as they can be observed as accurately as possible, they can be used to determine the variables. This can reduce the impact of disturbances on the motor position control error.

[0033] In one implementation, reference Figure 2 As shown, the above method also includes the step of obtaining the preset linear cascaded perturbation observer: Step S210: Construct an augmented state space model between the state variables and the total disturbance, and based on the augmented state space model, construct an N-level cascaded first-order linear disturbance observer, where N is a positive integer.

[0034] The augmented state-space model is used to represent the dynamic relationship between state variables and total disturbance. From equation (5), it can be seen that when u... d u q i q When used as input, i can be d i q w e The corresponding differential equations can be viewed as first-order systems, thus obtaining i d i q w e Augmented state-space models corresponding to the total disturbance.

[0035] With the state variable being the d-axis current i d For example, let's define it as follows: x1 = i d x2=f d u=u d b0=1 / L s And define g=g d =df d / dt, which is the bounded augmented form of the disturbance differential, the d-axis current equation in equation (5) can be expressed as an augmented state-space model between the d-axis current and the total disturbance, as shown below: (9) Among them, in equation (9) A, B, C, D and state variables x They are defined as follows: (10) Similarly, for the q-axis current i q Define x1=i q x2=f q u=u q b0=1 / L s and define g q =df q / dt, where f q G represents the total disturbance corresponding to the q-axis current. q This represents the rate of change of the total disturbance with respect to time. Therefore, the differential equation corresponding to the q-axis current can be expressed as an augmented state-space model with the same form as equation (9).

[0036] For motor speed ω e Define x1=ω e x2=f ω u=i q b0=3p²λ f / 2J, and define g w =df ω / dt, where f ω g represents the total disturbance corresponding to the rotational speed. w This represents the rate of change of the total disturbance with respect to time. The i in the speed equation of equation (5) q Using the corresponding coefficients as input gain b0, the augmented state-space model between rotational speed and total disturbance can be obtained.

[0037] Therefore, the d-axis current, q-axis current, and rotational speed can all be uniformly expressed as the first-order augmented system shown in equation (9), and corresponding linear cascaded disturbance observers can be constructed respectively.

[0038] An N-stage cascaded first-order linear perturbation observer refers to N first-order linear cascaded perturbation observers connected in series. For a first-order system, that is, for an augmented state-space model, constructing an N-stage cascaded first-order linear perturbation observer can be expressed as the following equation: (11) In the formula, For the corresponding observer The estimated disturbance, For the corresponding observer The estimated disturbance, Applying equations (9) and (11) above, we have: The symbol “^” represents the observed value of the corresponding state variable. and For observer Gain.

[0039] Step S220: Discretize the N-stage series first-order linear cascade perturbation observer to obtain the preset linear cascade perturbation observer.

[0040] Discretization refers to discretization using a first-order truncated Taylor expansion method. The pre-defined linear cascaded perturbation observer obtained after discretization is shown below, where T s The sampling interval is in seconds (s). (12) This disclosure constructs an augmented state-space model between state variables and total disturbance, and designs an N-stage series first-order linear cascade disturbance observer to treat internal disturbance changes and external load changes as a unified total disturbance. This structured design avoids the cumbersome parameter identification process, reduces the computational complexity, and enables centralized observation of multi-source disturbances. In addition, discretization further simplifies the real-time calculation of the observer, and can balance disturbance observation accuracy and noise suppression capability, thereby improving the robustness of motor position control.

[0041] In one implementation, reference Figure 3As shown, the above-mentioned linear cascade disturbance observer, based on a preset model, observes the total disturbance currently experienced by the permanent magnet synchronous motor, obtaining the disturbance observation value at the current moment, including: Step S310: Input the state variables and control signals of the permanent magnet synchronous motor at the current moment into the preset linear cascade disturbance observer, and use the first-level first-order linear cascade disturbance observer to observe and obtain the first-level observation value. Step S320: Input the first-level observation value into the second-level first-order linear cascade perturbation observer for observation, and so on, until the perturbation observation value output by the Nth-level first-order linear cascade perturbation observer is obtained.

[0042] Each first-order linear cascaded perturbation observer observes the total perturbation at the current moment and outputs an observation value for that level, which is then passed as input to the next first-order linear cascaded perturbation observer for further observation, continuously refining the observation values ​​until the last level, i.e., the Nth level, outputs the final perturbation observation value.

[0043] Based on equation (12) above, the perturbation observation error of each first-order linear cascade perturbation observer can be derived as a function of the perturbation transfer function G. e (s) is: (13) Where s is the complex frequency domain variable in the Laplace transform (different from the subscript s in equations (1)-(4) above); N Let be the total number of series-connected first-order linear perturbation observers. ω o ( ω o >0) is the observer bandwidth parameter of the linear cascaded perturbation observer, and s in the following transfer functions has the same meaning.

[0044] It can be seen that the values ​​of each gain parameter are: For a structure consisting of N first-order linear cascaded perturbation observers connected in series, to simplify parameter tuning, we take: (14) Based on the values ​​in equation (14), when N first-order linear cascaded perturbation observers are connected in series, the error transfer function G in equation (13) is... e (s) can be expressed as: (15) In one implementation, the gain parameters of the N-stage cascaded first-order linear perturbation observers are all identical; the value of N is 2.

[0045] When determining the value of N, in order to analyze the observation capability of the first-order linear cascade perturbation observer for perturbations, the perturbation following function G of the first-order linear cascade perturbation observer is used. f (s) can be expressed as follows, where the meaning of s is the same as that of the aforementioned error transfer function G. e The s in (s) are the same: (16) Table 2 compares the perturbation following function and error propagation function of the first-order linear cascade perturbation observer under different values ​​of the cascade number N.

[0046] Table 2. Error transfer function and disturbance follower function of LCDO under different series numbers N

[0047] As shown in Table 2, the more series connections there are, the more complex the observer needs to be designed and the greater the computational load. To balance observation accuracy and computational burden, N=2 is chosen.

[0048] This disclosure employs a multi-level, cascaded observation mechanism to gradually approximate the true perturbation value, reducing the impact of single observation errors on the system. This hierarchical processing structure enhances the observer's ability to track time-varying perturbations (such as instantaneous load changes when a humanoid robot grasps an object), ensuring the real-time performance and accuracy of perturbation observations while considering computational burden.

[0049] Continue to refer to Figure 1 In step 130, based on the disturbance observation, the current state variable and the control signal, the predicted state variable of the permanent magnet synchronous motor at the next moment is calculated.

[0050] The predicted state variables include predicted current and predicted rotational speed. The predicted current includes d-axis current and q-axis current. When predicting the d-axis current at the next moment, the state variables at the current moment include the d-axis current and the voltage is the d-axis reference voltage. Similarly, when predicting the q-axis current at the next moment, the state variables at the current moment include the q-axis current and the voltage is the q-axis reference voltage.

[0051] The predicted state variables for the next time step are not only affected by the current state variables and control signals, but also by the disturbance observations, making the predicted state variables more adaptable to the current application scenario and more robust.

[0052] In one implementation, reference Figure 4 The above-mentioned calculation of the predicted state variables of the permanent magnet synchronous motor at the next moment, based on the disturbance observation, the current state variables, and the control signal, includes: Step S410: Construct the motor state equation of the permanent magnet synchronous motor, and discretize the motor state equation to obtain the motor state prediction model. Step S420: Input the disturbance observation value, the current state variable and the control signal into the motor state prediction model to obtain the predicted state variable of the permanent magnet synchronous motor at the next moment.

[0053] The motor state equation is equation (5) mentioned above. Since equation 5 is a continuous-time prediction model, in order to accurately predict the next moment, equation 5 is discretized to obtain the following motor state prediction model. Through the following motor state prediction model, the d-axis current, q-axis current, and speed at the next moment can be predicted: (17) (18) (19) Where k is the current time and k+1 is the next time. Equation (11) defines... , which means the differential of the total disturbance during rotational speed.

[0054] In one example, discretization can be achieved using the truncated Taylor expansion method.

[0055] This disclosure constructs a prediction model by discretizing the motor state equations. By directly inputting the disturbance observations into the state variable prediction process, it corrects the model deviation that may be caused by parameter mismatch, making the predicted state variable at the next moment closer to the actual operating state of the motor. This provides a high-precision calculation basis for the subsequent reference voltage calculation, fundamentally improving the accuracy of position control.

[0056] Continue to refer to Figure 1 In step S140, the reference voltage of the permanent magnet synchronous motor is determined based on the predicted state variable and the preset target current at the next moment, so that the permanent magnet synchronous motor can perform position control on the robot based on the reference voltage at the next moment.

[0057] The reference voltage of a permanent magnet synchronous motor includes the reference voltages for the d-axis and q-axis.

[0058] In one implementation, reference Figure 5 The determination of the reference voltage of the permanent magnet synchronous motor based on the predicted current at the next moment and the preset target current includes: Step S510: Based on the current state variable, the disturbance observation value, and the preset target current, construct the deadbeat control formula for the reference voltage; Step S520: Substitute the predicted state variable and the disturbance observation value at the next moment into the deadbeat control formula to calculate the reference voltage of the permanent magnet synchronous motor at the next moment.

[0059] The preset target current includes the d-axis target current i d and q-axis target current i q The target current is generated by the robot's upper-level position control loop based on the deviation between the target position and the current actual position. Specifically, the upper-level position controller first generates corresponding torque or speed control commands, which are then further converted into target currents for the d-axis and q-axis of the permanent magnet synchronous motor. Therefore, i d and i q These represent the d-axis target current and q-axis target current corresponding to the current position control task, respectively.

[0060] First, without considering delay compensation, the deadbeat control formula for the reference voltage is determined. Based on the above equations (17) and (18), the following can be obtained: (20) (twenty one) Among them, u d (k) represents the d-axis reference voltage calculated without considering delay compensation, u q (k) is the q-axis reference voltage calculated without considering delay compensation.

[0061] Based on the truncated Taylor expansion method, the position equation of the permanent magnet synchronous motor can be discretized as: (twenty two) Where k represents the current discrete sampling time, and k+1 represents the next discrete sampling time. To maintain consistency of notation throughout the paper, the discrete-time model uniformly uses k and k+1 to represent two adjacent sampling times; in the continuous-time model, t only represents the continuous-time variable, T p The discrete period of the position.

[0062] Based on the above equations (20) and (21), and combined with the concept of deadbeat control, the reference voltage of the permanent magnet synchronous motor at the next moment is... for: (twenty three) (twenty four) Because this disclosure employs a structure combining observation and prediction, it uses the observed values ​​of state variables in the observer. As a predicted value in delay compensation, the reference voltage is therefore used when considering delay compensation. for: (25) (26) in, This represents the total disturbance estimate of the d-axis current at the next sampling time, which is obtained recursively from the discretized linear cascaded disturbance observer according to equation (12); correspondingly, and These represent the total disturbance estimates of the q-axis current and rotational speed at the next sampling time, respectively.

[0063] This disclosure further refines the reference voltage generation mechanism, introduces the concept of deadbeat control, and combines it with delay compensation. By substituting the disturbance observation value and predicted state variable into the calculation formula, a delay compensation alternative combining observation and prediction is realized. This further reduces the error accumulation of traditional one-beat delay compensation when parameters are mismatched, ensuring that the reference voltage applied at the next moment can accurately drive the permanent magnet synchronous motor to the target position, and significantly reducing the position control error.

[0064] In one implementation, the state variables of the permanent magnet synchronous motor at the next moment obtained by the motor state prediction model include the predicted current and the predicted speed. After determining the reference voltage of the permanent magnet synchronous motor at the next moment based on the predicted state variables and the preset target current, the reference voltage is then used. Figure 6a The method further includes: Step S610: Obtain the preset maximum current value and maximum speed value; Step S620: Using the preset maximum current value and maximum speed value, limit the predicted current and the predicted speed to obtain a safe current value and a safe speed value, and calculate the safe threshold of the reference voltage based on the safe current value and the safe speed value. Step S630: If the determined reference voltage exceeds the safety threshold, the reference voltage is truncated to obtain the reference voltage of the permanent magnet synchronous motor at the next moment.

[0065] The safety threshold includes the minimum value u of the d-axis reference voltage. d1 (k+1) and the maximum value u d2 (k+1), minimum value of q-axis reference voltage u q1 (k+1) and the maximum value u q2 (k+1), in addition, it also includes the minimum voltage boundary value u at a specific rotational speed. w1 (k+1) and the maximum value u w2 (k+1), calculated as follows: (27) (28) (29) In the above formula (29) It can be approximated as: .

[0066] Among them, I dmax I represents the maximum current along the d-axis. qmax This indicates the maximum current along the q-axis. This represents the maximum rotational speed. Since the above formula yields three safety threshold ranges, the final safety threshold is the intersection of these three ranges, thus simultaneously satisfying current constraints, rotational speed constraints, and corresponding voltage boundary constraints.

[0067] If the calculated reference voltage for the next moment does not exceed the safety threshold, then the subsequent processing is carried out according to the calculated reference voltage. If the calculated reference voltage for the next moment exceeds the safety threshold, then the reference voltage is truncated, and the exceeded safety threshold is taken as the reference voltage for the permanent magnet synchronous motor at the next moment. For example, if the calculated reference voltage is 250V and the maximum value in the final safety threshold is 220V, then 220V is taken as the reference voltage for the next moment.

[0068] This disclosure transforms current and speed constraints into voltage boundaries by performing safety threshold judgment and truncation processing after calculating the reference voltage, and uses saturation limiting to achieve constraint management. This prevents the risk of voltage over-limit caused by constraint boundary offset due to parameter mismatch, and ensures that the robot will not be damaged by overcurrent due to overcurrent or affect motion stability due to overspeed when encountering sudden load impact, thus significantly improving the safety and reliability of the robot.

[0069] In one implementation, when calculating the reference voltage, current and speed constraints can be directly incorporated as inequality constraints into the quadratic programming problem. Although this increases the computational load, it improves the robustness of constraint handling. To reduce online computation time, the piecewise affine (PWA) solution for explicit MPC can be calculated offline and stored in a lookup table. This approach ensures that the current does not exceed limits even when motor parameters vary widely, offering superior safety compared to the saturation limiting method.

[0070] In one implementation, to further reduce the computational burden, a first-order linear disturbance observer with N=1 can be used. In this case, the formula for calculating the reference voltage is simplified to: (30) Where x1 is the estimated value of current or rotational speed, x2 is the estimated value of total disturbance, and β1 and β2 are the observer gains. This first-order observer has a simpler structure and requires less parameter tuning, but its tracking speed for time-varying disturbances is slightly slower than the N=2 scheme. Under conditions where parameters change slowly or the load impact is not severe (such as slow walking or light-load operation of a humanoid robot), the steady-state position error can still be controlled within 0.5 rad, meeting general accuracy requirements.

[0071] In summary, the embodiments disclosed herein focus on predicting the state variables of permanent magnet synchronous motors and then using this prediction for robot position control. Their beneficial effects closely align with the needs of theoretical research and engineering applications, as detailed below: It exhibits outstanding robustness by capturing parameter fluctuations and load disturbances in real time through a linear cascaded disturbance observer, compensating for prediction models and constraint boundaries, and adapting to complex working conditions of humanoid robots. It boasts high control accuracy, with a structure combining observation and prediction replacing traditional delay compensation, reducing parameter mismatch errors and meeting high-precision positioning requirements. It is also highly user-friendly, continuing single-loop control and unweighted design to reduce parameter debugging. Furthermore, it offers broad adaptability, balancing dynamic response and steady-state performance, and can be expanded to multi-motor collaboration, perfectly meeting the core requirements of humanoid robot joint servo systems.

[0072] The above theory will be verified through simulation experiments: Given the position function: .

[0073] like Figure 6b The position curves and tracking error curves under no-load conditions are presented for a three-loop proportional-integral control (PI) and model predictive current control (MPCC) strategy. Analysis of the tracking error curves reveals that the maximum position error under the three-loop PI control is 0.844 rad at t=0.019s, while the maximum position error under MPCC is 0.1721 rad at t=0.002s. Here, SVPWM (Space Vector Pulse Width Modulation) refers to space vector pulse width modulation. dc Indicates the DC bus voltage, i a (k), i b (k), i c (k) represents the three-phase stator current (i.e., the currents of phase A, phase B, and phase C) at time k, and S1~S6 represent the six power switching transistors in the three-phase voltage source inverter.

[0074] Figure 6cTo obtain the (a) position curve and (b) tracking error curve based on the three-loop PI control and MPCC control strategies, with the load being 0.1 Nm at t=0s and the load changing to 0.3 Nm at t=0.15s, the tracking error curve (b) is analyzed: the maximum position error under the three-loop PI control is 0.8458 rad at t=0.019s, and the maximum position error under the MPCC control is 0.3312 rad at t=0.298s.

[0075] This disclosure also provides a robot position control device based on motor current prediction, with reference to... Figure 7 As shown, the robot position control device 700 based on motor current prediction includes: The information acquisition module 710 is used to acquire the state variables and control signals of the permanent magnet synchronous motor at the current moment during the process of the permanent magnet synchronous motor continuously controlling the position of the robot. The disturbance observation module 720 is used to observe the total disturbance currently experienced by the permanent magnet synchronous motor based on a preset linear cascade disturbance observer, and obtain the disturbance observation value at the current moment. The current prediction module 730 is used to calculate the predicted state variables of the permanent magnet synchronous motor at the next moment based on the disturbance observation value, the current state variables and control signals; The position control module 740 is used to determine the reference voltage of the permanent magnet synchronous motor at the next moment based on the predicted state variable and the preset target current at the next moment, so that the permanent magnet synchronous motor can perform position control on the robot based on the reference voltage at the next moment.

[0076] In one embodiment, the device further includes: The observer construction module is used to construct an augmented state space model between the state variables and the total disturbance, and based on the augmented state space model, construct an N-level cascaded first-order linear cascaded disturbance observer, where N is a positive integer; The observer acquisition module is used to discretize the N-stage series first-order linear cascade perturbation observer to obtain the preset linear cascade perturbation observer.

[0077] In one embodiment, the disturbance observation module 720 is specifically used for: The state variables and control signals of the permanent magnet synchronous motor at the current moment are input to the preset linear cascade disturbance observer, and the first-level first-order linear cascade disturbance observer is used to observe and obtain the first-level observation value. The first-level observation is input into the second-level first-order linear cascade perturbation observer for observation, and so on, until the perturbation observation output by the Nth-level first-order linear cascade perturbation observer is obtained.

[0078] In one implementation, The gain parameters of the N-stage cascaded first-order linear perturbation observers are all the same; the value of N is 2.

[0079] In one embodiment, the current prediction module 730 is specifically used for: The motor state equation of the permanent magnet synchronous motor is constructed, and the motor state equation is discretized to obtain the motor state prediction model. The disturbance observation, the current state variable, and the control signal are input into the motor state prediction model to obtain the predicted state variable of the permanent magnet synchronous motor at the next moment.

[0080] In one embodiment, the position control module is specifically used for: Based on the current state variables, the disturbance observations, and the preset target current, a deadbeat control formula for the reference voltage is constructed. Substitute the predicted state variables and the disturbance observation values ​​at the next moment into the deadbeat control formula to calculate the reference voltage of the permanent magnet synchronous motor at the next moment.

[0081] In one embodiment, the state variables of the permanent magnet synchronous motor at the next moment obtained by the motor state prediction model include predicted current and predicted speed; the device further includes: The preset value acquisition module is used to acquire the preset maximum current value and maximum speed value; The safety threshold calculation module is used to limit the predicted current and the predicted speed using the preset maximum current value and maximum speed value to obtain a safe current value and a safe speed value, and to calculate the safety threshold of the reference voltage based on the safe current value and the safe speed value. A voltage cutoff module is used to cut off the reference voltage when the determined reference voltage exceeds the safety threshold, so as to obtain the reference voltage of the permanent magnet synchronous motor at the next moment.

[0082] The specific details of each part of the above-mentioned device have been described in detail in the method section of the implementation plan. For any undisclosed details, please refer to the implementation plan of the method section, and therefore will not be repeated here.

[0083] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to exemplary embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0084] This disclosure also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the method steps of various exemplary embodiments of this disclosure.

[0085] In one implementation, the computer program product can be a tangible product, such as a computer-readable storage medium storing a computer program. The readable storage medium can be based on electrical, magnetic, optical, electromagnetic, infrared, or other signals, and includes, but is not limited to: Random Access Memory (RAM), Read-Only Memory (ROM), magnetic tape, floppy disk, flash memory, Hard Disk Drive (HDD), Solid State Disk (SSD), etc. For example, the computer program product can be a non-volatile storage medium storing a computer program, such as read-only memory, NAND flash memory, etc.

[0086] In one implementation, the computer program product can be an intangible product. For example, the computer program product can be a virtual digital product, such as an executable file or installation package containing a computer program.

[0087] Computer program code can be written in one or more programming languages. Examples of programming languages ​​include C, Java, and C++. Program code can execute entirely on the user's computing device, partially on the user's computing device, or as a standalone software package. It can also execute partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, such as a Local Area Network (LAN) or a Wide Area Network (WAN), or it can be connected to an external computing device (e.g., via an internet connection provided by a mobile network operator).

[0088] Computer programs can be carried or transmitted via signals such as electrical, magnetic, optical, electromagnetic, and infrared rays. Electronic devices can convert the signals carrying computer programs into digital signals, thereby running the computer programs. When a computer program runs on an electronic device, its code is used to cause the electronic device to execute (more specifically, to be executed by the processor of the electronic device) the method steps of various embodiments of this disclosure, such as... Figure 1 The method and steps.

[0089] Implementing the above method steps through a computer program achieves the following technical effects: This disclosure utilizes a linear cascaded disturbance observer to observe the current disturbance experienced by the permanent magnet synchronous motor in real time, and incorporates the disturbance observation value into the prediction of the state variable at the next moment. Based on this, the reference voltage of the permanent magnet synchronous motor at the next moment is determined, realizing the correction of the prediction result according to the actual disturbance. By introducing the disturbance observation value, the prediction of the state variable at the next moment and the reference voltage no longer depend solely on fixed, known parameters, reducing the impact of parameter mismatch on position control accuracy. This allows the permanent magnet synchronous motor servo system to maintain high-precision position control even when parameters change dynamically, better meeting the fine motion requirements of humanoid robots.

[0090] This disclosure solves the problems in related technologies, such as the inability to perceive and compensate for disturbances in real time, resulting in lag in dynamic response, speed drops, and limb tremors. It meets the high reliability and high precision motion requirements of robots in smart water pumping stations and is better able to adapt to the complex and varied operating conditions of humanoid robots in smart water pumping stations. While improving position control accuracy, it also enhances the robustness, dynamic performance, and safety of humanoid robots, especially adapting to position control scenarios using multi-motor collaborative permanent magnet synchronous motor servo systems. It addresses the shortcomings of related technologies in terms of accuracy, disturbance rejection, safety, and collaborative control in humanoid robot applications, providing core technical support for the development of high-performance humanoid robot joint servo systems.

[0091] This disclosure also provides an electronic device. The electronic device includes a processor and a memory. The memory stores executable instructions for the processor, such as computer programs. The processor executes the executable instructions to perform the method steps of various exemplary embodiments of this disclosure.

[0092] The following is for reference. Figure 8 The electronic device is illustrated by way of a general-purpose computing device. It should be understood that... Figure 8 The electronic device 800 shown is merely an example and should not be construed as limiting the functionality or scope of this disclosure.

[0093] like Figure 8 As shown, the electronic device 800 may include: a processor 810, a memory 820, a bus 830, an I / O (input / output) interface 840, and a network adapter 850.

[0094] The memory 820 may include volatile memory, such as RAM 821 and cache unit 822, and may also include non-volatile memory, such as ROM 823. The memory 820 may also include one or more program modules 824, including but not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. For example, program module 824 may include the modules described above.

[0095] The processor 810 may include one or more processing units, such as an AP (Application Processor), a modem processor, a GPU (Graphics Processing Unit), an ISP (Image Signal Processor), a controller, an encoder, a decoder, a DSP (Digital Signal Processor), a baseband processor, and / or an NPU (Neural-Network Processing Unit).

[0096] The processor 810 can be used to execute executable instructions stored in the memory 820 to perform method steps of various embodiments of this disclosure, such as... Figure 1 The method and steps.

[0097] By executing the above method steps through processor 810, the following technical effects are achieved: This disclosure utilizes a linear cascaded disturbance observer to observe the disturbance currently received by the permanent magnet synchronous motor in real time, and incorporates the disturbance observation value into the prediction of the state variable at the next moment. Based on this, the reference voltage of the permanent magnet synchronous motor at the next moment is determined, realizing the correction of the prediction result according to the actual disturbance. By introducing the disturbance observation value, the prediction of the state variable at the next moment and the reference voltage no longer depend solely on fixed, known parameters, reducing the impact of parameter mismatch on position control accuracy. This allows the permanent magnet synchronous motor servo system to maintain high-precision position control even when parameters change dynamically, better meeting the fine motion requirements of humanoid robots.

[0098] This disclosure solves the problems in related technologies, such as the inability to perceive and compensate for disturbances in real time, resulting in lag in dynamic response, speed drops, and limb tremors. It meets the high reliability and high precision motion requirements of robots in smart water pumping stations and is better able to adapt to the complex and varied operating conditions of humanoid robots in smart water pumping stations. While improving position control accuracy, it also enhances the robustness, dynamic performance, and safety of humanoid robots, especially adapting to position control scenarios using multi-motor collaborative permanent magnet synchronous motor servo systems. It addresses the shortcomings of related technologies in terms of accuracy, disturbance rejection, safety, and collaborative control in humanoid robot applications, providing core technical support for the development of high-performance humanoid robot joint servo systems.

[0099] Bus 830 is used to connect different components of electronic device 800 and may include data bus, address bus and control bus.

[0100] Electronic device 800 can communicate with one or more external devices 900 (such as keyboard, mouse, external controller, etc.) through I / O interface 840.

[0101] Electronic device 800 can communicate with one or more networks via network adapter 850. For example, network adapter 850 can provide mobile communication solutions such as 3G / 4G / 5G, or wireless communication solutions such as wireless LAN, Bluetooth, and near-field communication. Network adapter 850 can communicate with other modules of electronic device 800 via bus 830.

[0102] In one embodiment, the electronic device 800 further includes a display for displaying a graphical user interface.

[0103] although Figure 8 As not shown in the diagram, other hardware and / or software modules may also be configured in the electronic device 800, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (Redundant Arrays of Independent Disks) systems, tape drives, and data backup storage systems.

[0104] As can be seen from the above, the technical solutions disclosed herein can be implemented as methods, apparatus, systems, computer program products, storage media, electronic devices, etc. Those skilled in the art will understand that various aspects of this disclosure can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects. Exemplarily, these three forms can be referred to as "circuit," "module," and "system," respectively.

[0105] It should be understood that this disclosure is not limited to the specific methods, steps, or structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. Those skilled in the art will readily conceive of other embodiments based on the specific implementations provided in this disclosure. Therefore, the specific implementations provided in this disclosure are merely exemplary, and the scope and spirit of this disclosure are indicated by the claims, and should cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary technical means in the art not disclosed in this disclosure.

Claims

1. A robot position control method based on motor current prediction, characterized in that, include: During the continuous position control of the robot by the permanent magnet synchronous motor, the state variables and control signals of the permanent magnet synchronous motor at the current moment are acquired. Based on a preset linear cascade disturbance observer, the total disturbance currently experienced by the permanent magnet synchronous motor is observed to obtain the disturbance observation value at the current moment. Based on the disturbance observation, the current state variables, and the control signal, calculate the predicted state variables of the permanent magnet synchronous motor at the next moment; Based on the predicted state variables and the preset target current at the next moment, the reference voltage of the permanent magnet synchronous motor at the next moment is determined, so that the permanent magnet synchronous motor can perform position control on the robot based on the reference voltage at the next moment.

2. The method according to claim 1, characterized in that, The method further includes: An augmented state-space model is constructed between the state variables and the total disturbance, and based on the augmented state-space model, an N-level cascaded first-order linear cascaded disturbance observer is constructed, where N is a positive integer; Discretize the N-stage cascaded first-order linear perturbation observer to obtain the preset linear cascaded perturbation observer.

3. The method according to claim 2, characterized in that, The preset linear cascade disturbance observer observes the total disturbance currently experienced by the permanent magnet synchronous motor, obtaining the disturbance observation value at the current moment, including: The state variables and control signals of the permanent magnet synchronous motor at the current moment are input to the preset linear cascade disturbance observer, and the first-level first-order linear cascade disturbance observer is used to observe and obtain the first-level observation value. The first-level observation is input into the second-level first-order linear cascade perturbation observer for observation, and so on, until the perturbation observation output by the Nth-level first-order linear cascade perturbation observer is obtained.

4. The method according to claim 3, characterized in that, The gain parameters of the N-stage cascaded first-order linear perturbation observers are all the same; the value of N is 2.

5. The method according to claim 1, characterized in that, The calculation of the predicted state variables of the permanent magnet synchronous motor at the next moment based on the disturbance observation, the current state variables, and the control signal includes: The motor state equation of the permanent magnet synchronous motor is constructed, and the motor state equation is discretized to obtain the motor state prediction model. The disturbance observation, the current state variable, and the control signal are input into the motor state prediction model to obtain the predicted state variable of the permanent magnet synchronous motor at the next moment.

6. The method according to claim 5, characterized in that, The step of determining the reference voltage of the permanent magnet synchronous motor at the next moment based on the predicted state variable and the preset target current at the next moment includes: Based on the current state variables, the disturbance observations, and the preset target current, a deadbeat control formula for the reference voltage is constructed. Substitute the predicted state variables and the disturbance observation values ​​at the next moment into the deadbeat control formula to calculate the reference voltage of the permanent magnet synchronous motor at the next moment.

7. The method according to claim 1, characterized in that, The predicted state variables of the permanent magnet synchronous motor at the next moment include the predicted current and the predicted speed. After determining the reference voltage of the permanent magnet synchronous motor at the next moment based on the predicted state variable and the preset target current at the next moment, the method further includes: Obtain the preset maximum current value and maximum speed value; Using the preset maximum current value and maximum speed value, the predicted current and the predicted speed are limited to obtain a safe current value and a safe speed value, and the safety threshold of the reference voltage is calculated based on the safe current value and the safe speed value. If the determined reference voltage exceeds the safety threshold, the reference voltage is truncated to obtain the reference voltage of the permanent magnet synchronous motor at the next moment.

8. A robot position control device based on motor current prediction, characterized in that, include: The information acquisition module is used to acquire the state variables and control signals of the permanent magnet synchronous motor at the current moment during the continuous position control of the robot by the permanent magnet synchronous motor. The disturbance observation module is used to observe the total disturbance currently experienced by the permanent magnet synchronous motor based on a preset linear cascade disturbance observer, and obtain the disturbance observation value at the current moment. The current prediction module is used to calculate the predicted state variables of the permanent magnet synchronous motor at the next moment based on the disturbance observation value, the state variables at the current moment, and the control signal. The position control module is used to determine the reference voltage of the permanent magnet synchronous motor at the next moment based on the predicted state variable and the preset target current, so that the permanent magnet synchronous motor can perform position control on the robot at the next moment based on the reference voltage.

9. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to execute the method of any one of claims 1 to 7 by executing the executable instructions.