A deep-sea robot drive motor anti-interference cubic sliding mode position control method

Abstract

This invention discloses a disturbance-resistant cubic sliding mode position control method for a deep-sea robot drive motor. The method includes acquiring the actual operating parameters of the deep-sea robot drive motor and reconstructing the generalized lumped disturbance; estimating the generalized lumped disturbance of the motor under deep-sea operating conditions in real time; constructing a cubic nonlinear sliding mode surface containing linear and cubic error terms; designing a sliding mode reaching law based on a saturation limiting function and performing feedforward compensation in conjunction with the generalized lumped disturbance estimate to calculate the final q-axis current control command; and determining the robust gain coefficient of the lumped disturbance observer and the switching gain condition of the sliding mode controller based on Lyapunov stability theory. This invention effectively solves the problem of position tracking performance degradation caused by parameter perturbations and load mutations under complex deep-sea conditions. Without adding a traditional low-pass filter, it effectively overcomes steady-state phase lag and improves the dynamic response speed, tracking accuracy, and robustness of the drive motor position servo system.

Description

Technical Field

[0001] This invention relates to the field of motor control technology for deep-sea robots, and more specifically, to a method for anti-disturbance cubic sliding mode position control of a drive motor for a deep-sea robot. Background Technology

[0002] In recent years, with the rapid development of industry and intelligent manufacturing, permanent magnet synchronous motors have become the preferred choice for servo drive systems of high-end equipment such as deep-sea robots due to their high power density, high efficiency, and excellent dynamic response performance. The dynamic response speed and steady-state tracking accuracy of deep-sea robot operating equipment directly determine the quality of deep-sea exploration and operations.

[0003] However, in the complex and harsh operating environment of the deep sea, the position servo system of the drive motor of deep-sea robots often faces severe challenges: such as sudden changes in rotational inertia caused by the robotic arm grasping heavy objects, nonlinear friction in the transmission mechanism, and unknown external load disturbances brought by deep ocean currents. Traditional proportional-integral-derivative control, due to its fixed parameters, is prone to overshoot if the gain is increased in pursuit of faster response speed; conversely, if the gain is reduced for smooth operation, it will lead to severe phase lag and steady-state error when the system tracks continuous trajectories.

[0004] Sliding mode control is widely used due to its strong robustness, but traditional linear or higher-order sliding mode control often faces singularity or chattering problems. To enhance the system's disturbance rejection capability, existing technologies often introduce nonlinear observers or extended state observers. However, these traditional observers typically employ discontinuous sign functions (sgn) to pursue fast estimation, which inevitably introduces high-frequency noise. Therefore, existing systems are often forced to use a series low-pass filter (LPF) to eliminate noise, but this inevitably introduces phase delay, severely degrading the tracking performance of the motor position system.

[0005] Therefore, a new solution is needed to address the above problems. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for anti-disturbance cubic sliding mode position control of a deep-sea robot drive motor.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A method for disturbance-resistant cubic sliding mode position control of a drive motor for a deep-sea robot includes the following steps:

[0009] Step S1: Obtain the actual operating parameters of the deep-sea robot drive motor, and reconstruct the generalized lumped disturbance, which includes system parameter perturbation and load disturbance, based on motor dynamics.

[0010] Step S2: Construct a lumped disturbance observer based on the hyperbolic tangent function, and estimate the generalized lumped disturbance of the motor under deep-sea operation conditions in real time according to the rotor angular velocity estimation error.

[0011] Step S3: Based on the rotor position tracking error, construct a cubic nonlinear sliding surface containing linear and cubic error terms to provide adaptive adjustment capability for variable damping.

[0012] Step S4: Design a sliding mode reaching law based on the saturation limiting function, and perform feedforward compensation by combining the generalized lumped disturbance estimate to calculate the final q-axis current control command to drive the deep-sea robot motor to run at the target position;

[0013] Step S5: Based on Lyapunov stability theory, determine the robust gain coefficient of the lumped disturbance observer and the switching gain condition of the sliding mode controller to ensure the asymptotic stability of the entire position control system.

[0014] Furthermore, in step S1, the reconstruction process of the generalized lumped disturbance includes:

[0015] Step S11: Establish the nominal system dynamic equations of the motor position servo system;

[0016] Step S12: Express the actual moment of inertia and viscous friction coefficient as the sum of the nominal value and the disturbance value, and construct the generalized lumped disturbance.

[0017] Step S13: Based on the generalized lumped disturbance, nominal moment of inertia, and nominal viscous friction coefficient, reconstruct the nominal system dynamic equation.

[0018] Furthermore, in step S2, the process of constructing the lumped disturbance observer includes:

[0019] Step S21: Define the rotor angular velocity estimate, and combine the nominal moment of inertia, nominal viscous friction coefficient, actual rotor angular velocity, estimate of generalized lumped disturbance and corresponding q-axis current to establish a velocity observer model based on the dynamic equation of the motor position system.

[0020] Step S22: Define the rotor angular velocity estimation error, introduce the robust gain coefficient and linear width parameter, and design a nonlinear error injection law based on the hyperbolic tangent function;

[0021] Step S23: Based on the nonlinear error injection law, establish the closed-loop error dynamic equation.

[0022] Furthermore, in step S3, the construction process of the cubic nonlinear sliding surface includes:

[0023] Step S31: Define the position tracking error based on the current rotor reference position and the actual position;

[0024] Step S32: Based on the position tracking error, linear term coefficients, and cubic term coefficients, establish the cubic nonlinear sliding surface equation;

[0025] Step S33: Differentiate the cubic nonlinear sliding surface equation to obtain the dynamic equation of the sliding variable, and define the damping coefficient used to adjust the system damping amplitude in real time according to the error magnitude.

[0026] Furthermore, in step S4, the specific process of designing the sliding mode reaching law based on the saturation limiting function includes:

[0027] Step S41: Based on the switching gain, linear feedback gain, and boundary layer thickness, establish a reaching law with saturation limiting characteristics;

[0028] Step S42, construct the saturation limiting function Its definition is:

[0029] when hour, The system state rapidly approaches the sliding surface driven by a fixed switching gain;

[0030] when hour, This transforms the control into pure linear feedback control, eliminating high-frequency switching oscillations.

[0031] Furthermore, in step S4, the process of calculating the final q-axis current control command includes:

[0032] Step 1: Substitute the nominal system dynamic equation into the dynamic equation of the sliding variable, and multiply both sides of the equation by the nominal moment of inertia.

[0033] Step 2: Ensure the system dynamically and strictly follows the designed saturation limit approach law;

[0034] Step 3: Use the estimated value output by the lumped disturbance observer to perform equivalent feedforward compensation for the unknown generalized lumped disturbance, and design the final quadrature-axis current command control law.

[0035] Furthermore, in step S5, the condition for determining the robust gain coefficient of the lumped disturbance observer based on Lyapunov stability theory is as follows:

[0036] Constructing a Lyapunov function for the angular velocity estimation error Due to the generalized lumped disturbance in deep-sea operations Bounded, satisfied Select robust gain coefficient ,satisfy To ensure the derivative This ensures that the observer system meets stability requirements; among which, Generalized aggregate disturbance The upper bound; This represents the rotor angular velocity estimation error.

[0037] Furthermore, in step S5, the condition for determining the switching gain of the sliding mode controller based on Lyapunov stability theory is as follows:

[0038] Define the residual estimation error of the observer. Constructing the Lyapunov function of the system Select switching gain ,satisfy ,make sure This holds true at all times, ensuring the global asymptotic stability of the sliding mode control system after the observer rapidly tracks and compensates for the disturbance; among which, For generalized aggregate disturbance; This is an estimate of the generalized lumped disturbance; This is the time derivative of the Lyapunov function; This represents the linear feedback gain in the sliding mode reaching law; It is a sliding surface.

[0039] Furthermore, the method also includes: when the deep-sea robot's drive motor is running, the actual three-phase current is transformed into d / q-axis feedback current through coordinate transformation, the q-axis feedback current and rotor angular velocity are input into the lumped disturbance observer, the reference position, actual position, actual angular velocity and the estimated value of generalized lumped disturbance are input into the cubic nonlinear sliding mode controller, the q-axis current command is calculated and output, and after current inner loop PI regulation, coordinate inverse transformation and voltage modulation, the inverter is driven so that the motor runs at the target position.

[0040] The beneficial effects of this invention are:

[0041] 1. This invention designs a lumped disturbance observer based on the hyperbolic tangent function to address the abrupt changes in rotational inertia, nonlinear friction, and unknown ocean current loads encountered by deep-sea robots under complex operating conditions. Simultaneously, it smooths the switching process of the discontinuous sign function sgn in traditional observers and suppresses high-frequency noise. Accurate estimation and feedforward compensation of lumped disturbances are achieved without the need for a traditional low-pass filter (LPF) in series.

[0042] 2. This invention constructs a cubic nonlinear sliding surface containing linear and cubic error terms. When the deep-sea robot arm grasps a heavy object and generates a large initial position error, the system is dominated by the cubic term, providing a strong nonlinear restoring force and driving the system state to converge rapidly. When the error approaches zero, the cubic term decays rapidly and is dominated by the highly damped linear term, ensuring that the system smoothly enters a steady state. This perfectly overcomes the huge overshoot problem caused by pursuing response speed in traditional proportional-integral control.

[0043] 3. This invention replaces the sign function approach law in traditional sliding mode control with a saturation limiting function. When the system state is far from the sliding surface, a fixed gain is used for rapid approach. Once inside the sliding boundary layer, the system automatically and smoothly transitions to pure linear feedback control. This mechanism fundamentally eliminates high-frequency switching chattering in the control current. It avoids high-frequency wear on motor drive gears and mechanical joints, significantly enhancing the robustness of the position servo system.

[0044] 4. The method of the present invention is not only applicable to deep-sea robot electric drive systems, but can also be extended to electric vehicles, industrial servo systems, and wind power generation. Attached Figure Description

[0045] Figure 1 This is a topology diagram of the anti-disturbance cubic sliding mode position control system for the deep-sea robot drive motor in this embodiment;

[0046] Figure 2 This is a block diagram of a lumped perturbation observer based on the hyperbolic tangent function in this embodiment;

[0047] Figure 3 This is a control logic flowchart of the anti-disturbance cubic sliding mode position control method for the drive motor of the deep-sea robot in this embodiment;

[0048] Figure 4 The waveform diagram shows the response test comparison between the method of the present invention and traditional proportional-integral control and conventional linear sliding mode control when the drive motor of the deep-sea robot undergoes a step change at the target position in this embodiment.

[0049] Figure 5 The waveform diagram shows the comparison of the tracking test between the method of the present invention and traditional proportional-integral control and conventional linear sliding mode control under the continuous sinusoidal trajectory condition of the deep-sea robot drive motor in this embodiment.

[0050] Figure 6 The waveform diagram shows the dynamic tracking test comparison between the method of the present invention and traditional proportional-integral control and conventional linear sliding mode control under the trapezoidal acceleration and deceleration conditions of the deep-sea robot drive motor in this embodiment. Detailed Implementation

[0051] 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.

[0052] Example: A disturbance-resistant cubic sliding mode position control system for a deep-sea robot drive motor, such as... Figure 1As shown, the deep-sea robot's drive motor outputs three-phase current during operation. , , The actual rotor mechanical angle is obtained by the speed / position extraction module. With mechanical angular velocity The three-phase current is first transformed into a stationary coordinate system current through the abc→αβ transformation. , Combined with mechanical angles The d / q axis feedback current in the rotating coordinate system is obtained through the αβ→dq transformation. , Among them, feedback current With mechanical angular velocity Input the lumped disturbance observer and calculate the output generalized lumped disturbance estimate. Meanwhile, reference position Mechanical angular velocity Mechanical angle and lumped disturbance estimate The "cubic nonlinear sliding mode control method" module is used as a common input to calculate and output the q-axis current command. And set the d-axis current command. =0; subsequently, the d-axis and q-axis current errors are adjusted by the PI controller to output d-axis and q-axis voltage commands respectively. , Next, the voltage command. , Combining mechanical perspective The voltage in the stationary coordinate system is obtained by the inverse transformation from dq to αβ. , The input voltage modulation module generates the inverter switching signal. Ultimately, the inverter operates based on the switching signal. DC bus voltage The voltage is inverted to a three-phase AC voltage, which drives the motor to run at the target position, completing the closed-loop control cycle.

[0053] A disturbance-resistant cubic sliding mode position control method for a deep-sea robot drive motor addresses the degradation in tracking performance and response speed of the position servo system caused by abrupt changes in rotational inertia, nonlinear friction, and external unknown load disturbances under complex seabed conditions. First, a lumped disturbance observer based on the hyperbolic tangent function is constructed to perform real-time smooth estimation of the generalized lumped disturbance of the system, including inertia mismatch, using actual angular velocity and quadrature-axis current. Second, based on the position tracking error, a nonlinear sliding surface combining cubic and linear terms is introduced to endow the system with variable damping characteristics, balancing rapid convergence under large errors and smooth transition under small errors. Then, by combining the saturation limiting reaching law and disturbance feedforward compensation from the observer output, a smooth quadrature-axis current command is calculated and output, completely eliminating the need for traditional low-pass filters and the resulting phase lag. Finally, the real-time updated current command is used for high-precision position tracking without overshoot or chattering. This method suppresses parameter perturbations and load disturbances in the deep-sea environment, improves the dynamic response capability of the drive motor servo system, and enhances the overall robustness of the motor system.

[0054] like Figures 1-3 As shown, the method includes the following steps:

[0055] Step S1: Obtain the actual operating parameters of the deep-sea robot's drive motor, and reconstruct the generalized lumped disturbance, including system parameter perturbations and load disturbances, based on motor dynamics. Specifically:

[0056] Step S11, establish the nominal system dynamic equations of the motor position servo system:

[0057]

[0058] In the formula, Let the system's rotational inertia be denoted by . , , These are the rotor mechanical angle, angular velocity, and angular acceleration, respectively. This is the torque coefficient; This is the q-axis current; It is the coefficient of viscous friction; For unknown load torque in the deep sea; It is a nonlinear frictional torque.

[0059] Step S12: Considering mechanical wear and physical parameter perturbations in the deep-sea environment, define the actual moment of inertia. With viscous friction coefficient It is the sum of the nominal value and the disturbance value, i.e. and Constructing a generalized aggregate disturbance ;

[0060]

[0061] In the formula, The nominal moment of inertia; The nominal coefficient of viscous friction; This represents the disturbance value of the moment of inertia. This represents the disturbance value of the viscous friction coefficient.

[0062] Step S13, based on generalized lumped disturbance Nominal moment of inertia Nominal viscous friction coefficient Reconstruct the nominal system dynamic equations:

[0063]

[0064] In the formula, This is the derivative of the actual angular velocity of the rotor; This is the actual angular velocity of the rotor. .

[0065] Further, in step S2, a lumped disturbance observer is constructed based on the hyperbolic tangent function. Based on the rotor angular velocity estimation error, the generalized lumped disturbance of the motor under deep-sea operating conditions is estimated in real time, such as... Figure 2 As shown:

[0066] Step S21, establish a velocity observer model based on the dynamic equations of the motor position system:

[0067]

[0068] In the formula, The derivative of the estimated rotor angular velocity; This is an estimate of the generalized lumped disturbance.

[0069] Step S22, define the rotor angular velocity estimation error. To avoid high-frequency chattering caused by traditional sign functions, a robust gain coefficient is introduced. With linear width parameter A smoothing nonlinear error injection law is designed using the hyperbolic tangent function:

[0070]

[0071] In the formula, and ; , This is the actual angular velocity of the rotor. This is an estimated value for the rotor angular velocity.

[0072] Step S23: Substitute the nonlinear error injection law designed in step S22 into the system to establish the closed-loop error dynamic equation:

[0073]

[0074] In the formula, This is the derivative of the rotor angular velocity estimation error.

[0075] Furthermore, in step S3, based on the rotor position tracking error, a cubic nonlinear sliding surface containing linear and cubic error terms is constructed to provide adaptive adjustment capability for variable damping, such as... Figure 3 As shown;

[0076] Step S31, based on the current rotor reference position With actual location Define position tracking error :

[0077]

[0078] Step S32, based on position tracking error Linear term coefficients Cube term coefficient Establish the equations for the cubic nonlinear sliding surface:

[0079]

[0080] In the formula, and ; It is a sliding surface; For position tracking error The first derivative.

[0081] The position-current dual closed-loop control system for the deep-sea robot's drive motor possesses "variable damping characteristics": when the absolute value of the position tracking error... When large, the cubic term Dominant, providing powerful nonlinear restoring force to drive rapid error convergence; when position tracking error Approaching zero ( )hour, The terms decay rapidly, and the sliding surface changes from linear terms. It dominates and ensures an extremely smooth steady state.

[0082] Step S33: Differentiate the cubic nonlinear sliding surface equation to obtain the dynamic equation of the sliding variable, and then define the damping coefficient. , Used to adjust the system damping amplitude in real time according to the magnitude of the error:

[0083]

[0084] In the formula, The derivative of the sliding surface; For position tracking error The first derivative; For position tracking error The second derivative; For reference position The second derivative, i.e., the reference angular acceleration; The actual mechanical angle of the rotor The second derivative of , i.e., the actual angular acceleration; and These are the coefficients of the linear term and the coefficients of the cubic term, respectively.

[0085] Step S4: Design a sliding mode reaching law based on the saturation limiting function, and perform feedforward compensation by combining the generalized lumped disturbance estimate output in step S2 to calculate the final q-axis current control command to drive the deep-sea robot motor to run at the target position.

[0086] Step S41, establish a reaching law with saturation limiting characteristics:

[0087]

[0088] In the formula, To switch the gain, >0; For linear feedback gain, >0; For the specified boundary layer thickness, >0.

[0089] Step S42, construct the saturation limiting function Its definition is:

[0090] When the system state is far from the sliding surface, i.e. hour, The system state rapidly approaches the sliding surface driven by a fixed switching gain;

[0091] When the system state enters the boundary layer, that is... hour, This transforms the control into pure linear feedback control, eliminating high-frequency switching oscillations.

[0092] In step S4, the process of calculating the final q-axis current control command includes:

[0093] Step 1: Substitute the nominal system dynamic equation into the dynamic equation of the sliding variable, and multiply both sides of the equation by the nominal moment of inertia. The calculation formula is:

[0094]

[0095] In the formula, For reference position The second derivative; Damping coefficient; This is the torque coefficient; The nominal coefficient of viscous friction; This is the actual angular velocity of the rotor; The nominal moment of inertia; For generalized aggregate disturbance; This is the q-axis current; For position tracking error The first derivative; is the derivative of the sliding surface.

[0096] Step 2, ensure the system dynamically and strictly follows the designed saturation limit approach law, that is, let , The saturation limit approach law for the design;

[0097] Step 3: Use the estimated values ​​output by the lumped disturbance observer. Unknown generalized lumped disturbance Equivalent feedforward compensation is performed, and the final q-axis current command control law is designed. :

[0098]

[0099] Step S5: Based on Lyapunov stability theory, determine the robust gain coefficient of the lumped disturbance observer and the switching gain condition of the sliding mode controller to ensure the asymptotic stability of the entire position control system.

[0100] The condition for determining the robust gain coefficient of the lumped perturbation observer based on Lyapunov stability theory is as follows:

[0101] Constructing a Lyapunov function for the angular velocity estimation error Due to the generalized lumped disturbance in deep-sea operations Bounded, satisfied Select robust gain coefficient ,satisfy To ensure the derivative This ensures that the observer system meets stability requirements; among which, Generalized aggregate disturbance The upper bound; This represents the rotor angular velocity estimation error.

[0102] The condition for determining the switching gain of the sliding mode controller based on Lyapunov stability theory is as follows:

[0103] Define the residual estimation error of the observer. Constructing the Lyapunov function of the system Select switching gain ,satisfy ,make sure This holds true at all times, ensuring the global asymptotic stability of the sliding mode control system after the observer rapidly tracks and compensates for the disturbance; among which, For generalized aggregate disturbance; This is an estimate of the generalized lumped disturbance; This is the time derivative of the Lyapunov function; This represents the linear feedback gain in the sliding mode reaching law; It is a sliding surface.

[0104] The effectiveness of the method proposed in this embodiment is verified below:

[0105] Figure 4 The waveforms show a comparison of the response tests of the traditional proportional-integral (PI) control method, the conventional linear sliding mode control method, and the disturbance-resistant cubic sliding mode control method of this embodiment when the position jumps from 0 rad to 1 rad. The traditional PI control strategy, due to its fixed parameters, produces a large overshoot (peak value approximately 1.68 rad) when pursuing fast response, and has a relatively long settling time of approximately 0.28 s. While the conventional linear sliding mode control strategy effectively eliminates overshoot, its rise time is approximately 0.07 s due to the use of a fixed linear sliding surface gain, resulting in a certain response delay. In contrast, the proposed disturbance-resistant cubic sliding mode control strategy... The strong restoring force provided allows the actual position to approach the reference command at the fastest speed, reducing the step rise time to 0.035s. When the error approaches zero, the linearly damped system smoothly enters steady state, avoiding overshoot. In addition, the saturation limiting approach law effectively suppresses switching chattering in the steady-state phase.

[0106] Figure 5 The waveforms show a comparison of the traditional proportional-integral (PI) control method, the conventional linear sliding mode control method, and the disturbance-resistant cubic sliding mode control method proposed in this embodiment, under continuous sinusoidal trajectory tracking test conditions. From the waveforms and magnified views, it is clear that the traditional PI control method exhibits approximately 8% amplitude attenuation at the peak, accompanied by a phase lag of approximately 0.04 s, and significant motor noise at the peak. The conventional linear sliding mode control mitigates the noise to some extent and achieves a smooth transition at the peak, but it still has a dynamic tracking error of 0.02 rad. In contrast, the actual running trajectory of the proposed disturbance-resistant cubic sliding mode control almost perfectly matches the reference sine curve, with the maximum tracking error remaining within 0.005 rad. This indicates that the proposed error injection law based on the hyperbolic tangent function achieves smooth disturbance estimation and can perform feedforward compensation for continuously changing nonlinear friction and lumped disturbances without delay, ensuring high-precision position tracking under time-varying trajectories.

[0107] Figure 6 The diagram shows a comparison of the dynamic tracking waveforms of the traditional proportional-integral (PI) control method, the conventional linear sliding mode control method, and the disturbance-resistant cubic sliding mode position control method proposed in this embodiment under trapezoidal acceleration and deceleration conditions. At the inflection point between system acceleration and deceleration at the platform, the instantaneous electromagnetic torque output by the system is extremely high. As shown in the magnified area, the traditional PI control method, due to integral lag, completely loses its dynamic tracking capability at the inflection point, exhibiting severe trajectory rounding and a maximum dynamic error of approximately 0.24 rad, recovering to steady state after approximately 2.30 seconds. While the conventional linear sliding mode control possesses some robustness, its thrust is limited at the moment of sudden acceleration change, also resulting in an error of approximately 0.06 rad and a recovery delay. In contrast, the proposed disturbance-resistant cubic sliding mode control strategy effectively overcomes the dynamic trajectory tracking lag at abrupt corners by using a lumped disturbance observer for feedforward compensation of disturbances and combining the disturbance resistance capability of the cubic nonlinear sliding surface under large errors. The dynamic error remains consistently within a very small range of 0.004 rad, demonstrating stronger robustness.

[0108] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for disturbance-resistant cubic sliding mode position control of a deep-sea robot drive motor, characterized in that, Includes the following steps: Step S1: Obtain the actual operating parameters of the deep-sea robot drive motor, and reconstruct the generalized lumped disturbance, which includes system parameter perturbation and load disturbance, based on motor dynamics. Step S2: Construct a lumped disturbance observer based on the hyperbolic tangent function, and estimate the generalized lumped disturbance of the motor under deep-sea operation conditions in real time according to the rotor angular velocity estimation error. Step S3: Based on the rotor position tracking error, construct a cubic nonlinear sliding surface containing linear and cubic error terms to provide adaptive adjustment capability for variable damping. Step S4: Design a sliding mode reaching law based on the saturation limiting function, and perform feedforward compensation by combining the generalized lumped disturbance estimate to calculate the final q-axis current control command to drive the deep-sea robot motor to run at the target position; Step S5: Based on Lyapunov stability theory, determine the robust gain coefficient of the lumped disturbance observer and the switching gain condition of the sliding mode controller to ensure the asymptotic stability of the entire position control system.

2. The method for anti-disturbance cubic sliding mode position control of a deep-sea robot drive motor according to claim 1, characterized in that, In step S1, the reconstruction process of the generalized lumped disturbance includes: Step S11: Establish the nominal system dynamic equations of the motor position servo system; Step S12: Express the actual moment of inertia and viscous friction coefficient as the sum of the nominal value and the disturbance value, and construct the generalized lumped disturbance. Step S13: Based on the generalized lumped disturbance, nominal moment of inertia, and nominal viscous friction coefficient, reconstruct the nominal system dynamic equation.

3. The method for anti-disturbance cubic sliding mode position control of a deep-sea robot drive motor according to claim 1, characterized in that, In step S2, the process of constructing the lumped disturbance observer includes: Step S21: Define the rotor angular velocity estimate, and combine the nominal moment of inertia, nominal viscous friction coefficient, actual rotor angular velocity, estimate of generalized lumped disturbance and corresponding q-axis current to establish a velocity observer model based on the dynamic equation of the motor position system. Step S22: Define the rotor angular velocity estimation error, introduce the robust gain coefficient and linear width parameter, and design a nonlinear error injection law based on the hyperbolic tangent function; Step S23: Based on the nonlinear error injection law, establish the closed-loop error dynamic equation.

4. The method for anti-disturbance cubic sliding mode position control of a deep-sea robot drive motor according to claim 1, characterized in that, In step S3, the construction process of the cubic nonlinear sliding surface includes: Step S31: Define the position tracking error based on the current rotor reference position and the actual position; Step S32: Based on the position tracking error, linear term coefficients, and cubic term coefficients, establish the cubic nonlinear sliding surface equation; Step S33: Differentiate the cubic nonlinear sliding surface equation to obtain the dynamic equation of the sliding variable, and define the damping coefficient used to adjust the system damping amplitude in real time according to the error magnitude.

5. The method for anti-disturbance cubic sliding mode position control of a deep-sea robot drive motor according to claim 1, characterized in that, In step S4, the specific process of designing the sliding mode reaching law based on the saturation limiting function includes: Step S41: Based on the switching gain, linear feedback gain, and boundary layer thickness, establish a reaching law with saturation limiting characteristics; Step S42, construct the saturation limiting function Its definition is: when hour, The system state rapidly approaches the sliding surface driven by a fixed switching gain; when hour, This transforms the control into pure linear feedback control, eliminating high-frequency switching oscillations.

6. The method for anti-disturbance cubic sliding mode position control of a deep-sea robot drive motor according to claim 5, characterized in that, In step S4, the process of calculating the final q-axis current control command includes: Step 1: Substitute the nominal system dynamic equation into the dynamic equation of the sliding variable, and multiply both sides of the equation by the nominal moment of inertia. Step 2: Ensure the system dynamically and strictly follows the designed saturation limit approach law; Step 3: Use the estimated value output by the lumped disturbance observer to perform equivalent feedforward compensation for the unknown generalized lumped disturbance, and design the final quadrature-axis current command control law.

7. The method for anti-disturbance cubic sliding mode position control of a deep-sea robot drive motor according to claim 1, characterized in that, In step S5, the condition for determining the robust gain coefficient of the lumped perturbation observer based on Lyapunov stability theory is as follows: Constructing a Lyapunov function for the angular velocity estimation error Due to the generalized lumped disturbance in deep-sea operations Bounded, satisfied Select robust gain coefficient ,satisfy To ensure the derivative This ensures that the observer system meets stability requirements; among which, Generalized aggregate disturbance The upper bound; This represents the rotor angular velocity estimation error.

8. The method for anti-disturbance cubic sliding mode position control of a deep-sea robot drive motor according to claim 1, characterized in that, In step S5, the condition for determining the switching gain of the sliding mode controller based on Lyapunov stability theory is as follows: Define the residual estimation error of the observer. Constructing the Lyapunov function of the system Select switching gain ,satisfy ,make sure This holds true at all times, ensuring the global asymptotic stability of the sliding mode control system after the observer rapidly tracks and compensates for the disturbance; among which, For generalized aggregate disturbance; This is an estimate of the generalized lumped disturbance; This is the time derivative of the Lyapunov function; This represents the linear feedback gain in the sliding mode reaching law; It is a sliding surface.

9. The method for anti-disturbance cubic sliding mode position control of a deep-sea robot drive motor according to claim 1, characterized in that, The method also includes: when the deep-sea robot's drive motor is running, the actual three-phase current is transformed into d / q-axis feedback current through coordinate transformation, the q-axis feedback current and rotor angular velocity are input into the lumped disturbance observer, the reference position, actual position, actual angular velocity and the estimated value of generalized lumped disturbance are input into the cubic nonlinear sliding mode controller, the q-axis current command is calculated and output, and after current inner loop PI regulation, coordinate inverse transformation and voltage modulation, the inverter is driven so that the motor runs at the target position.

Images