An underwater robot dynamic positioning method for complex disturbance

Abstract

The application provides an underwater robot dynamic positioning method for complex disturbance, which generates fused high-frequency high-precision position information by collecting the attitude angle, angular velocity, linear velocity and linear acceleration information of the underwater robot; designs an extended state disturbance observer, estimates the disturbance value in real time through a feedback function and an auxiliary variable equation; adds the disturbance estimation value to the output force of the dynamics model to construct the constraint condition of model predictive control, solves the optimal control input of the future N prediction time domains through the rolling horizon optimization; based on the optimal control input, combined with the spatial position distribution matrix of the thruster, constructs a quadratic programming problem of thrust distribution, solves the minimum energy solution of the thrust of each thruster, and outputs to the thruster actuator. The technical scheme of the application can realize the robust control of the underwater robot in a strong disturbance environment. The application mainly solves the fixed-point and fixed-orientation problem of the underwater unmanned robot when facing underwater turbulent flow or other complex disturbance.

Description

Technical Field

[0001] This invention belongs to the field of autonomous control technology for unmanned underwater robots, and specifically relates to a dynamic positioning method for underwater robots to cope with complex disturbances. Background Technology

[0002] In recent years, the complexity of underwater robot operations and scenarios has been increasing, leading to several technical challenges, particularly in motion control. These challenges include: firstly, the high degree of uncertainty in the underwater environment, with disturbances such as undercurrents and ocean currents; and secondly, the problem of accurate path tracking in high-damped environments.

[0003] The dynamic coupling forces of underwater robots are complex, and the parameters of their dynamic models have significant uncertainties. For model-based control, dynamic models are currently the most widely used. Methods for establishing dynamic models mainly fall into two categories: knowledge-driven modeling and data-driven modeling. Knowledge-driven modeling involves constructing a spatial motion model forward based on Newton's laws of motion and mathematical physics methods. A commonly used model is the hydrodynamic model for AUVs and ROVs proposed by Fossen. This model can be further simplified based on the actual operating environment and payload configuration of the underwater robot.

[0004] Based on established models, internationally accepted control methods include model predictive control and sliding mode control. For state variables that are difficult to measure, such as modeling errors and external disturbances, model-based disturbance observers are often used to observe and estimate these control variables. Although underwater positioning of underwater robots can be achieved to some extent, problems such as inaccurate positioning, instability, and difficulty in implementation still exist. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a dynamic positioning method for underwater robots to cope with complex disturbances, which can achieve robust control of underwater robots in strong disturbance environments and solve the problem of fixed-point and fixed-attitude operation of underwater unmanned robots when facing underwater turbulence or other complex disturbances.

[0006] In this embodiment of the invention, a dynamic positioning method for an underwater robot to cope with complex disturbances is provided, comprising:

[0007] S101. The attitude angle, angular velocity, linear velocity and linear acceleration information of the underwater robot are collected by the inertial measurement unit (IMU) and the binocular camera. The error state Kalman filter algorithm is used to calibrate the gyroscope noise of the IMU, the accelerometer bias and the pose measurement noise of the binocular camera, and generate fused high-frequency and high-precision pose information.

[0008] S102. Based on the six-degree-of-freedom dynamic model of an underwater robot, an extended state disturbance observer (ESO) is designed to model external unknown disturbances and internal parameter changes as extended state variables, and the disturbance value is estimated in real time through feedback functions and auxiliary variable equations.

[0009] S103. Sum the disturbance estimate with the output force of the dynamic model to construct the constraints of model predictive control (MPC). Define the loss function with the weighted sum of squares of the state tracking error and the quadratic term of the control input. Solve for the optimal control input in the next N prediction time domains through rolling time domain optimization.

[0010] S104. Based on the optimal control input and combined with the spatial pose distribution matrix of the thrusters, a quadratic programming problem of thrust distribution is constructed with the maximum thrust of each thruster as the inequality constraint. The minimum energy solution of the thrust of each thruster is solved by the Lagrange multiplier method and output to the thruster actuator.

[0011] Furthermore, the implementation of the error state Kalman filter algorithm includes:

[0012] The random walk coefficient of the gyroscope angle of the IMU, the zero bias instability parameter of the accelerometer, and the image feature point matching noise variance of the binocular camera were calibrated by experiment, and the calibration parameters were written into the YAML format configuration file.

[0013] A state-space equation is constructed with attitude quaternion error, velocity error and position error as state variables. Its state transition matrix includes the angular velocity integral term of the IMU and the gravity compensation term of the accelerometer.

[0014] The Kalman gain matrix is ​​dynamically adjusted based on the sampling frequency of the binocular camera to ensure that the frequency of the fused pose data is not lower than the original sampling frequency of the IMU.

[0015] Furthermore, the design of the extended state perturbation observer satisfies the following conditions:

[0016] Assume that the external disturbance force has weak time-varying properties within the sampling period, that is, the first derivative of the disturbance force with time is approximately zero;

[0017] Define the auxiliary variable as a linear combination of the observer's state variables, whose dynamic equation satisfies in, As auxiliary variables, K1 and K2 are diagonal positive definite feedback matrices, and x is the actual state of the system. Estimate the state for the observer.

[0018] Furthermore, the loss function of the MPC controller is defined as:

[0019]

[0020] Where x(k) is the state vector at the k-th step in the prediction time domain, including position, velocity, and attitude angle; x ref (k) represents the desired state at the corresponding time, Q is the diagonal weight matrix, and its diagonal elements correspond to the penalty coefficients for position error, velocity error and attitude error; R is the diagonal weight matrix, which controls the coefficient of the quadratic penalty term of the input increment; Δu(k) controls the input increment to avoid drastic fluctuations.

[0021] Furthermore, the method also includes a dynamic disturbance compensation mechanism:

[0022] When the binocular camera loses pose data due to underwater turbidity, it switches to the short-term pose prediction mode of the pure inertial measurement unit (IMU) and compensates for the positioning error caused by the lack of visual data through the extended state perturbation observer (ESO).

[0023] Furthermore, based on the statistically weak time-varying assumption of the high-intensity flow field, the derivative of the perturbation force with time is approximated as 0, and an extended state perturbation observer is designed.

[0024] The specific design method of the ESO perturbation observer includes: taking the feedback function P(v) = L(v)v; assuming the auxiliary variable z satisfies Perturbation observer estimate It can be represented as

[0025] Tracking error e satisfies Based on the assumption that the perturbation force is weakly time-varying, the derivative of the perturbation force is considered to be 0. Designing L as a constant matrix, we can obtain It can be seen that it eventually converges to 0.

[0026] Furthermore, the system control force and external disturbance force are summed to obtain the resultant force that the actual thruster needs to output. This resultant force is then integrated into the constraints and loss function of the MPC to obtain a more physically meaningful optimal solution, specifically in the following form:

[0027] Introducing weight matrices Q and W, and the target state R, the following loss function is defined:

[0028]

[0029] in, To extend the position matrix, the constraints can be expressed as: Where λ is the scaling factor and F is the ultimate thrust in each direction; by minimizing the loss function, the control quantities for the next N predicted time domains are obtained. Each calculation only takes the control quantity at time k as the control input, thereby realizing the rolling optimization process.

[0030] Furthermore, the method also includes an energy consumption optimization strategy, namely:

[0031] The weighting coefficients of the Lagrange multiplier method are updated online based on historical thrust allocation data, prioritizing the reduction of the usage frequency of high-power thrusters;

[0032] When the battery level is detected to be below a threshold, the position tracking accuracy constraint is automatically relaxed to reduce total energy consumption and extend battery life.

[0033] Furthermore, the fusion localization process of the inertial measurement unit (IMU) includes:

[0034] The ORB-SLAM algorithm is used to construct an underwater environmental feature point map in real time, and the camera coordinate system is transformed to the IMU body coordinate system through the hand-eye calibration matrix;

[0035] By using the prediction-update loop of error state Kalman filtering, visual pose corrections are inserted between high-frequency IMU data to generate a fused pose output above 200Hz.

[0036] The beneficial effects of this invention are as follows:

[0037] As can be seen from the above scheme, the embodiments of the present invention provide a dynamic positioning method for underwater robots to cope with complex disturbances. By collecting the underwater robot's attitude angle, angular velocity, linear velocity, and linear acceleration information, the gyroscope noise of the IMU, the accelerometer bias, and the pose measurement noise of the binocular camera are calibrated to generate fused high-frequency, high-precision pose information. External unknown disturbances and internal parameter changes are modeled as extended state variables, and the disturbance value is estimated in real time through feedback functions and auxiliary variable equations. The disturbance estimate is summed with the output force of the dynamic model to construct constraints for model predictive control. The loss function is defined by the weighted sum of squares of the state tracking error and the quadratic term of the control input. The optimal control input for the next N prediction time domains is solved through rolling time-domain optimization. Based on the optimal control input, combined with the spatial pose distribution matrix of the thrusters, and using the maximum thrust of each thruster as an inequality constraint, a quadratic programming problem for thrust allocation is constructed to solve for the minimum energy solution of the thrust of each thruster, and the solution is output to the thruster actuator. The technical solution of the present invention can achieve robust control of underwater robots in strong disturbance environments. This invention primarily addresses the problem of maintaining a fixed position and attitude for underwater unmanned robots when facing underwater turbulence or other complex disturbances. Attached Figure Description

[0038] Figure 1 This is a flowchart of a dynamic positioning method for an underwater robot to cope with complex disturbances according to an embodiment of the present invention;

[0039] Figure 2This is a schematic diagram of the effect of an expanded state disturbance observer in an underwater robot dynamic positioning method for dealing with complex disturbances according to an embodiment of the present invention.

[0040] Figure 3 This is a schematic diagram comparing the dynamic positioning effect of an underwater robot dynamic positioning method for dealing with complex disturbances with traditional control methods, according to an embodiment of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0042] like Figures 1 to 3 As shown, Figure 1 This is a flowchart of a dynamic positioning method for an underwater robot to cope with complex disturbances, according to an embodiment of the present invention. Figure 2 This is a schematic diagram illustrating the effect of an expanded state disturbance observer in an underwater robot dynamic positioning method for dealing with complex disturbances, according to an embodiment of the present invention. Figure 3 This is a schematic diagram comparing the dynamic positioning effect of an underwater robot dynamic positioning method for dealing with complex disturbances with traditional control methods, according to an embodiment of the present invention.

[0043] Figure 1 Among them, a dynamic localization method for underwater robots to cope with complex disturbances includes:

[0044] S101. The attitude angle, angular velocity, linear velocity and linear acceleration information of the underwater robot are collected by the inertial measurement unit (IMU) and the binocular camera. The error state Kalman filter algorithm is used to calibrate the gyroscope noise of the IMU, the accelerometer bias and the pose measurement noise of the binocular camera, and generate fused high-frequency and high-precision pose information.

[0045] S102. Based on the six-degree-of-freedom dynamic model of an underwater robot, an extended state disturbance observer (ESO) is designed to model external unknown disturbances and internal parameter changes as extended state variables, and the disturbance value is estimated in real time through feedback functions and auxiliary variable equations.

[0046] S103. Sum the disturbance estimate with the output force of the dynamic model to construct the constraints of model predictive control (MPC). Define the loss function with the weighted sum of squares of the state tracking error and the quadratic term of the control input. Solve for the optimal control input in the next N prediction time domains through rolling time domain optimization.

[0047] S104. Based on the optimal control input and combined with the spatial pose distribution matrix of the thrusters, a quadratic programming problem of thrust distribution is constructed with the maximum thrust of each thruster as the inequality constraint. The minimum energy solution of the thrust of each thruster is solved by the Lagrange multiplier method and output to the thruster actuator.

[0048] In this invention, a dynamic positioning method for an unmanned underwater robot to cope with complex disturbances is presented. Based on IMU and binocular camera information, sensor data fusion is achieved through intrinsic parameter calibration, improving the accuracy and transmission frequency of the underwater robot's short-term pose and velocity data. The method introduces the assumption of weak time-varying statistical properties of the high-intensity flow field during the construction of the disturbance observer, enhancing the mathematical rigor of the convergence proof process. The MPC controller is used to calculate the control system input. Furthermore, the disturbance observer's observation results are integrated during this process to obtain a more physically meaningful optimal control force input. Using the Lagrange multiplier method on the calculation results of the MPC controller, thrust allocation is performed based on the spatial distribution of the thrusters, aiming to minimize energy consumption and improve the robot's endurance.

[0049] In one embodiment of the present invention, the implementation of the error state Kalman filter algorithm includes:

[0050] The random walk coefficient of the gyroscope angle of the IMU, the zero bias instability parameter of the accelerometer, and the image feature point matching noise variance of the binocular camera were calibrated by experiment, and the calibration parameters were written into the YAML format configuration file.

[0051] A state-space equation is constructed with attitude quaternion error, velocity error and position error as state variables. Its state transition matrix includes the angular velocity integral term of the IMU and the gravity compensation term of the accelerometer.

[0052] The Kalman gain matrix is ​​dynamically adjusted based on the sampling frequency of the binocular camera to ensure that the frequency of the fused pose data is not lower than the original sampling frequency of the IMU.

[0053] In one embodiment of the present invention, the design of the extended state perturbation observer satisfies the following condition:

[0054] Assume that the external disturbance force has weak time-varying properties within the sampling period, that is, the first derivative of the disturbance force with time is approximately zero;

[0055] Define the auxiliary variable as a linear combination of the observer's state variables, whose dynamic equation satisfies in, As auxiliary variables, K1 and K2 are diagonal positive definite feedback matrices, and x is the actual state of the system. Estimate the state for the observer.

[0056] In one embodiment of the present invention, the loss function of the MPC controller is defined as:

[0057]

[0058] Where x(k) is the state vector at the k-th step in the prediction time domain, including position, velocity, and attitude angle; x ref (k) represents the desired state at the corresponding time, Q is the diagonal weight matrix, and its diagonal elements correspond to the penalty coefficients for position error, velocity error and attitude error; R is the diagonal weight matrix, which controls the coefficient of the quadratic penalty term of the input increment; Δu(k) controls the input increment to avoid drastic fluctuations.

[0059] In one embodiment of the present invention, the method further includes a dynamic disturbance compensation mechanism:

[0060] When the binocular camera loses pose data due to underwater turbidity, it switches to the short-term pose prediction mode of the pure inertial measurement unit (IMU) and compensates for the positioning error caused by the lack of visual data through the extended state perturbation observer (ESO).

[0061] In one embodiment of the present invention, based on the statistical weak time-varying assumption of high-intensity flow field, the derivative of the perturbation force with time is approximated as 0, and an extended state perturbation observer is designed.

[0062] The specific design method of the ESO perturbation observer includes: taking the feedback function P(v) = L(v)v; assuming the auxiliary variable z satisfies Perturbation observer estimate It can be represented as

[0063] Tracking error e satisfies Based on the assumption that the perturbation force is weakly time-varying, the derivative of the perturbation force is considered to be 0. Designing L as a constant matrix, we can obtain It can be seen that it eventually converges to 0.

[0064] In one embodiment of the present invention, the system control force and the external disturbance force are summed to obtain the resultant force that the actual thruster needs to output. This resultant force is then integrated into the constraints and loss function of the MPC to obtain a more physically meaningful optimal solution. Specifically, the form is as follows:

[0065] Introducing weight matrices Q and W, and the target state R, the following loss function is defined:

[0066]

[0067] in, To extend the position matrix, the constraints can be expressed as: Where λ is the scaling factor and F is the ultimate thrust in each direction; by minimizing the loss function, the control quantities for the next N predicted time domains are obtained. Each calculation only takes the control quantity at time k as the control input, thereby realizing the rolling optimization process.

[0068] In one embodiment of the present invention, the method further includes an energy consumption optimization strategy, namely:

[0069] The weighting coefficients of the Lagrange multiplier method are updated online based on historical thrust allocation data, prioritizing the reduction of the usage frequency of high-power thrusters;

[0070] When the battery level is detected to be below a threshold, the position tracking accuracy constraint is automatically relaxed to reduce total energy consumption and extend battery life.

[0071] In one embodiment of the present invention, the fusion positioning process of the inertial measurement unit (IMU) includes:

[0072] The ORB-SLAM algorithm is used to construct an underwater environmental feature point map in real time, and the camera coordinate system is transformed to the IMU body coordinate system through the hand-eye calibration matrix;

[0073] By using the prediction-update loop of error state Kalman filtering, visual pose corrections are inserted between high-frequency IMU data to generate a fused pose output above 200Hz.

[0074] In this embodiment of the invention, the designed MPC controller integrates disturbance observation results. Unlike the method of directly subtracting the disturbance observation value from the MPC output control force, the observation results are utilized by adding the system control force and the external disturbance force to obtain the resultant force that the actual thruster needs to output, and integrating this resultant force into the constraints and loss function of the MPC to obtain a more physically meaningful optimal solution.

[0075] In one embodiment of the present invention, a dynamic positioning method for an unmanned underwater robot in response to complex disturbances is described in conjunction with the accompanying drawings.

[0076] First, the noise and random walk parameters of the IMU and stereo camera are calibrated and written to a YAML file. An error-state Kalman filter algorithm is used to fuse sensor information, and a state-space equation is constructed from acceleration, velocity, and attitude information. During data sampling, the Kalman gain is updated in real time based on the sampling frequency to improve sensor accuracy and data transmission frequency.

[0077] Based on the statistically weak time-varying assumption of high-intensity flow fields, the derivative of the perturbation force with time is approximated as 0, and an extended state perturbation observer is designed. The specific design method of the ESO perturbation observer is as follows: The feedback function is P(v) = L(v)v; the auxiliary variable z is assumed to satisfy... Perturbation observer estimate It can be represented as The proof is as follows: The tracking error e satisfies Based on the assumption that the perturbation force is weakly time-varying, the derivative of the perturbation force is considered to be 0. Designing L as a constant matrix, we can obtain It can be seen that it eventually converges to 0. For example... Figure 2 As shown, Figure 2 A schematic diagram illustrating the effect of the extended state perturbation observer used in this invention in a simulation environment is provided.

[0078] Based on the disturbance force observation results, an MPC controller is designed. Unlike the method of directly subtracting the disturbance observation value from the MPC output control force, the control system utilizes the observation results by summing the system control force with the external disturbance force to obtain the resultant force required for the actual thruster output. This resultant force is then integrated into the MPC constraints and loss function to obtain a more physically meaningful optimal solution. Specifically, by introducing weight matrices Q and W, and the target state R, the following loss function is defined. in To extend the position matrix, the constraints can be expressed as: Where λ is the scaling factor and F is the ultimate thrust in each direction. By minimizing the loss function, the control quantities for the next N predicted time domains can be obtained. Each calculation only takes the control quantity at time k as the control input, thereby realizing the rolling optimization process.

[0079] Based on the control input calculation results, the Lagrange multiplier method is used, with the ultimate thrust of the thrusters as the constraint condition and the sum of the squares of the thrusts of each thruster as the loss function. The thrust of each thruster is calculated and the results are applied to the control system to complete dynamic positioning control under disturbance conditions. Figure 3 As shown, Figure 3 This is a schematic diagram comparing the dynamic positioning effect of the present invention with that of traditional control methods.

[0080] This invention provides a dynamic positioning method for underwater robots to cope with complex disturbances. It collects the underwater robot's attitude angles, angular velocities, linear velocities, and linear accelerations, calibrates the IMU's gyroscope noise, accelerometer bias, and pose measurement noise of the binocular camera, and generates fused high-frequency, high-precision pose information. External unknown disturbances and internal parameter changes are modeled as extended state variables, and the disturbance values ​​are estimated in real time using feedback functions and auxiliary variable equations. The disturbance estimates are summed with the output force of the dynamic model to construct constraints for model predictive control. A loss function is defined using the weighted sum of squares of the state tracking error and the quadratic term of the control input. The optimal control input for the next N prediction time domains is solved through rolling time-domain optimization. Based on the optimal control input and the spatial pose distribution matrix of the thrusters, a quadratic programming problem for thrust allocation is constructed using the maximum thrust of each thruster as an inequality constraint. The minimum energy solution for the thrust of each thruster is solved and output to the thruster actuator.

[0081] The technical solution of this invention enables robust control of underwater robots in highly disturbed environments. This invention primarily addresses the problem of stationary orientation and fixation of underwater unmanned robots when facing underwater turbulence or other complex disturbances.

[0082] The above are preferred embodiments of the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A dynamic positioning method for underwater robots to cope with complex disturbances, characterized in that, The method includes: S101. The attitude angle, angular velocity, linear velocity and linear acceleration information of the underwater robot are collected by the inertial measurement unit (IMU) and the binocular camera. The error state Kalman filter algorithm is used to calibrate the gyroscope noise of the IMU, the accelerometer bias and the pose measurement noise of the binocular camera, and generate fused high-frequency and high-precision pose information. S102. Based on the six-degree-of-freedom dynamic model of an underwater robot, an extended state disturbance observer (ESO) is designed to model external unknown disturbances and changes in internal parameters as extended state variables, and the disturbance value is estimated in real time through feedback functions and auxiliary variable equations. S103. Summing the disturbance estimate with the output force of the dynamic model, constructing the constraints of Model Predictive Control (MPC), defining the loss function with the weighted sum of squares of the state tracking error and the quadratic term of the control input, and solving for the optimal control input for the next N prediction time domains through rolling time-domain optimization; summing the system control force with the external disturbance force to obtain the resultant force required for the actual thruster output, and integrating this resultant force into the constraints and loss function of MPC to obtain a more physically meaningful optimal solution, specifically in the following form: Introducing a weight matrix and and target state Define the following loss function: ， in, To extend the position matrix, the constraints can be expressed as: ,in, Scaling factor Given the ultimate thrust in each direction; by minimizing the loss function, the future thrust is obtained. Control quantity in the prediction time domain Each operation only takes The control quantity at any given time serves as the control input, thereby enabling the rolling optimization process; S104. Based on the optimal control input and combined with the spatial pose distribution matrix of the thrusters, a quadratic programming problem of thrust distribution is constructed with the maximum thrust of each thruster as the inequality constraint. The minimum energy solution of the thrust of each thruster is solved by the Lagrange multiplier method and output to the thruster actuator.

2. The underwater robot dynamic positioning method for coping with complex disturbances according to claim 1, characterized in that, The implementation of the error state Kalman filter algorithm includes: The random walk coefficient of the gyroscope angle of the IMU, the zero bias instability parameter of the accelerometer, and the image feature point matching noise variance of the binocular camera were calibrated by experiment, and the calibration parameters were written into the YAML format configuration file. A state-space equation is constructed with attitude quaternion error, velocity error and position error as state variables. Its state transition matrix includes the angular velocity integral term of the IMU and the gravity compensation term of the accelerometer. The Kalman gain matrix is ​​dynamically adjusted based on the sampling frequency of the binocular camera to ensure that the frequency of the fused pose data is not lower than the original sampling frequency of the IMU.

3. The underwater robot dynamic positioning method for coping with complex disturbances according to claim 1, characterized in that, The design of the extended state perturbation observer must satisfy the following conditions: Assume that the external disturbance force has weak time-varying properties within the sampling period, that is, the first derivative of the disturbance force with time is approximately zero; The auxiliary variable is defined as a linear combination of the observer's state variables, and its dynamic equation satisfies ,in, As an auxiliary variable, , It is a diagonal positive definite feedback matrix. This represents the actual state of the system. Estimate the state for the observer.

4. The underwater robot dynamic positioning method for coping with complex disturbances according to claim 1, characterized in that, The loss function of the MPC controller is defined as: in, To predict the first in the time domain The state vector of a step includes position, velocity, and attitude angle; This represents the desired state at the corresponding time. This is a diagonal weight matrix, where the diagonal elements correspond to the penalty coefficients for position error, velocity error, and attitude error. This is a diagonal weight matrix, controlling the coefficients of the quadratic penalty term for the input increment; Control the increment of input to avoid drastic fluctuations.

5. The underwater robot dynamic positioning method for coping with complex disturbances according to claim 1, characterized in that, The method also includes a dynamic disturbance compensation mechanism: When the binocular camera loses pose data due to underwater turbidity, it switches to the short-term pose prediction mode of the pure inertial measurement unit (IMU) and compensates for the positioning error caused by the lack of visual data through the extended state perturbation observer (ESO).

6. The underwater robot dynamic positioning method for coping with complex disturbances according to claim 1, characterized in that, Based on the statistical weak time-varying assumption of high-intensity flow field, the derivative of the perturbation force with time is approximated as 0, and an extended state perturbation observer is designed. The specific design method of the ESO perturbation observer includes: taking the feedback function. Let auxiliary variables be defined. satisfy Perturbation observer estimate It can be represented as ; Tracking error satisfy Based on the assumption that the perturbation force is weakly time-varying, the derivative of the perturbation force is considered to be 0. ;design As a constant matrix, we can obtain It can be seen that it eventually converges to 0.

7. The underwater robot dynamic positioning method for coping with complex disturbances according to claim 1, characterized in that, The method also includes an energy consumption optimization strategy, namely: The weighting coefficients of the Lagrange multiplier method are updated online based on historical thrust allocation data, prioritizing the reduction of the usage frequency of high-power thrusters; When the battery level is detected to be below a threshold, the position tracking accuracy constraint is automatically relaxed to reduce total energy consumption and extend battery life.

8. The underwater robot dynamic positioning method for coping with complex disturbances according to claim 1, characterized in that, The fusion positioning process of the inertial measurement unit (IMU) includes: The ORB-SLAM algorithm is used to construct an underwater environmental feature point map in real time, and the camera coordinate system is transformed to the IMU body coordinate system through the hand-eye calibration matrix; By using the prediction-update loop of error state Kalman filtering, visual pose corrections are inserted between high-frequency IMU data to generate a fused pose output above 200Hz.

Images