Amphibious vehicle anti-interference control method for optimizing ADRC based on H-infinity filtering and AFTESO

Abstract

The invention is suitable for the technical field of vehicle engineering and ship industry, and provides an amphibious vehicle anti-interference control method for optimizing ADRC based on H-infinity filtering and AFTESO, and the method comprises the steps: constructing a vehicle dynamics model, and calculating an expected motion state; performing filtering and disturbance preliminary identification on the rotating speed signal of the propeller by using an H-infinity filter; based on the error between the filtered state and the actual state, an adaptive finite time extended state observer (AFTESO) is adopted to accurately estimate the total disturbance of the system in real time; and in combination with an active-disturbance-rejection control (ADRC) algorithm, feedforward compensation is carried out on a disturbance estimation value to a control law, an anti-interference propeller rotating speed instruction is generated, and closed-loop control is formed. According to the method, multi-source coupling disturbance is effectively processed in a unified mode, the control precision, the attitude stability and the system robustness of the vehicle under the complex water navigation working condition are improved, and meanwhile dependence on an accurate model and hardware computing power is reduced.

Description

An Anti-interference Control Method for Amphibious Vehicles Based on H∞ Filtering and AFTESO Optimized ADRC Technical Field

[0001] This invention belongs to the fields of vehicle engineering and shipbuilding technology, and particularly relates to an anti-interference control method for amphibious vehicles based on H∞ filtering and AFTESO optimized ADRC. Background Technology

[0002] With the increasing demand for combined land and water transportation and emergency rescue, amphibious vehicles, especially heavy-duty transport vehicles, are facing increasingly widespread and complex applications. In water navigation conditions, vehicles simultaneously encounter external environmental disturbances such as wind, waves, and currents, as well as internal disturbances such as fluctuations in power system output. These multi-source, coupled disturbances severely affect the vehicle's navigation attitude stability and control accuracy, potentially leading to trajectory deviations or even safety accidents. Therefore, designing an anti-interference method capable of achieving precise and stable control in complex disturbance environments has become a key technical challenge in the field of amphibious vehicle engineering.

[0003] Currently, existing technical solutions for attitude and heading control of amphibious vehicles mainly fall into the following categories: first, methods that adjust the vehicle's attitude by regulating the speed difference between wheels or propellers; second, methods that use vehicle dynamics models combined with algorithms such as PID and Model Predictive Control (MPC) for steering or attitude control; and third, methods that use estimation techniques such as Kalman filters to filter specific disturbances (such as waves) and then combine them with traditional control laws. All these methods attempt to improve control performance under disturbed environments.

[0004] Despite the progress made in existing technologies, the following limitations still exist: First, the actuators of some methods have low efficiency, resulting in slow attitude adjustment response; second, many control strategies do not fully consider or integrate unified observation and compensation for multi-source disturbances (coupling of internal and external disturbances) during the design, resulting in insufficient robustness in complex disturbance environments; third, some high-performance control methods are highly dependent on the accurate mathematical model of the controlled object, and the online computation is large, requiring high hardware computing power, which increases the difficulty and cost of deployment in practical applications. Summary of the Invention

[0005] The purpose of this invention is to provide an anti-interference control method for amphibious vehicles based on H∞ filtering and AFTESO optimized ADRC, aiming to solve the technical problems existing in the prior art as identified in the background art.

[0006] This invention is implemented as follows: an anti-interference control method for amphibious vehicles based on H∞ filtering and AFTESO optimized ADRC, characterized by the following steps:

[0007] S1. Constructing the Vehicle Dynamics Model and Parameter Identification: Based on vehicle parameters and real-time data collected by sensors, a dynamic model of the amphibious heavy-duty transport vehicle is constructed, and the desired lateral and longitudinal velocities and the desired yaw angle are solved. Specifically, the vehicle parameters include the vehicle's length, width, height, and vehicle mass; the real-time data collected by sensors includes the vehicle's draft, the spray pump motor speed, and driver input information. Preferably, a three-degree-of-freedom vehicle dynamics model is constructed. This step also includes calculating the thrust and torque of the left and right propellers based on the external characteristic curve of the waterjet propeller, calculating the yaw resistance torque based on simplified engineering expressions, and identifying and verifying the model parameters through real vehicle data acquisition and the least squares method.

[0008] S2. Preliminary Disturbance Identification and Noise Suppression Based on H∞ Filter: Using the rotational speeds of the left and right thrusters as control inputs, a discretized state-space model is designed to encompass external disturbances such as wind, waves, and currents, as well as internal dynamic structure disturbances. Based on the H∞ filter, noise suppression and preliminary disturbance identification are performed on the rotational speed signals of the left and right thrusters, which are collected in real-time from vehicle speed feedback. The filter performance index γ is dynamically adjusted so that the difference between the actual yaw angle and the desired yaw angle meets a preset threshold, thereby obtaining the filtered desired state.

[0009] S3. Disturbance and State Estimation Based on AFTESO: Based on the error between the filtered desired state and the actual motion state, the finite-time extended state observer AFTESO is used to estimate the system state and total disturbance in real time. The dynamic equations of AFTESO are as described in the preceding claims.

[0010] S4. ADRC-based disturbance compensation and closed-loop control: Combining the active disturbance rejection control (ADRC) algorithm, the total disturbance value estimated by AFTESO is used for feedforward compensation of the control law to calculate the current propeller speed control command that can effectively suppress external and internal disturbances, and feed it back to the vehicle dynamics model as the control input to update the vehicle's force and motion state, thus forming a closed-loop control.

[0011] The beneficial effects of this invention are:

[0012] This invention combines the robust noise suppression and preliminary disturbance identification capabilities of the H∞ filter with the core concept of Active Disturbance Rejection Control (ADRC) and its accurate total disturbance estimation and compensation capabilities achieved through an Adaptive Finite-Time Extended State Observer (AFTESO), forming a collaborative anti-interference control architecture. This scheme can uniformly observe and effectively feedforward compensate for multiple external disturbances such as wind, waves, and currents encountered by amphibious vehicles during navigation, as well as internal system disturbances, without relying on a precise mathematical model of the controlled object. This significantly enhances the system's adaptability and robustness under strong disturbance conditions. Simultaneously, this scheme reduces the need for online optimization calculations, alleviates the computational burden on the controller, facilitates implementation in vehicle-mounted embedded systems, improves the engineering practicality and economy of the control strategy, and ultimately achieves rapid, accurate, and stable attitude control of amphibious heavy-duty transport vehicles in complex navigation environments. Attached Figure Description

[0013] Figure 1 is a structural diagram of the power system of an amphibious heavy-duty transport vehicle;

[0014] Figure 2 shows the anti-interference control technology scheme for amphibious heavy-duty transport vehicles.

[0015] Figure 3 is a flowchart of the H∞ filter control for an amphibious heavy-duty transport vehicle.

[0016] Figure 4 is a flowchart of the anti-interference control process for amphibious heavy-duty transport vehicles. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0018] The present invention provides an anti-interference control method for amphibious vehicles based on H∞ filtering and AFTESO optimized ADRC, which aims to improve the anti-interference capability and attitude stability of amphibious heavy-duty transport vehicles (whose power system structure can be referred to in Figure 1) under complex navigation conditions on water. The core control architecture is shown in Figure 2, mainly including a vehicle expected output calculation module and a disturbance identification and suppression module.

[0019] I. Vehicle Expected Output Calculation Module

[0020] By constructing a three-degree-of-freedom vehicle dynamics model (lateral, longitudinal, and yaw), the vehicle state equation is built based on information such as the vehicle's length, width, height, and body mass, combined with real-time data collected by sensors, including the vehicle's draft, injection pump motor speed, and driver input information (such as accelerator pedal, brake pedal, and gear position). This allows for the calculation of the desired lateral and longitudinal velocities and yaw angle, providing a data foundation for subsequent anti-interference control strategies.

[0021] When an amphibious vehicle is sailing smoothly on water, it can be considered a floating body, satisfying the condition that gravity and buoyancy are equal. However, the impact pressure of waves and the buoyancy fluctuations cause the hull to shift, thus disrupting the vehicle's original floating equilibrium and forcing non-uniform changes in the draft of different parts of the hull. Let the coordinates of the absolute center of mass in the inertial coordinate system be... After the hull undergoes offset and rotation, the relative coordinates of the center of mass in the hull coordinate system are obtained through rotation matrix transformation. This allows us to calculate the offset of the centroid position based on the offset of the coordinate axes.

[0022] Vehicle drag mainly includes frictional drag, form drag, and wave drag. Among these, the yaw drag torque is generated by the yaw motion of the vehicle about its vertical axis. The following simplified expression can be used to describe the project:

[0023] ;

[0024] In the formula: The vehicle's angular velocity; The drag coefficient was subsequently obtained through parameter identification using real vehicle data. These are the actual width, length, and height of the vehicle; The density of water is taken as the value at normal temperature and pressure. .

[0025] The thrust of an amphibious vehicle can be calculated based on the external characteristic curve of a waterjet propulsion system:

[0026] ;

[0027] ;

[0028] ;

[0029] In the formula: The static thrust coefficient at a constant thruster speed; The rotational speed of the thruster; This is a power arm for rotation.

[0030] The formula for yaw rate is:

[0031] ;

[0032] In the formula: They are respectively around The moment of inertia and additional moment of inertia of the shaft; data collected from the actual vehicle (around...) The system collects data on axle angular velocity, lateral and longitudinal vehicle speeds, and angular acceleration. It also performs time synchronization and outlier removal on the collected data to ensure data reliability. The system uses the least squares method to analyze the vehicle dynamics equations constructed in the vehicle dynamics model and solves for the optimal parameters to be identified. The validity of the parameters is confirmed by repeatability verification of multiple sets of data under the same working conditions.

[0033] II. Disturbance Identification and Suppression Module

[0034] This module is the core of anti-interference control, responsible for identifying multi-source disturbances and generating compensation control commands. Its process can be seen in Figure 4.

[0035] 1. Noise Suppression Based on H∞ Filter

[0036] The noise suppression module based on the H∞ filter acquires the rotational speed signals of the left and right thrusters in real time through the vehicle's rotational speed feedback, and receives the desired yaw angle command from the controller to provide data for subsequent noise suppression algorithms. Using the rotational speeds of the left and right thrusters as control inputs, a discretized state-space model is designed to encompass external disturbances such as wind, waves, and currents, as well as internal dynamic structural disturbances. By dynamically adjusting the filtering performance indicators, noise reduction and preliminary identification of system disturbances are achieved, improving the vehicle's heading control accuracy and navigation attitude stability in complex navigation environments.

[0037] As shown in Figure 3, the H∞ filter collects the propeller rotation speed signal of the amphibious heavy-duty transport vehicle and uses the desired yaw angle obtained from the vehicle dynamics model as the control input signal. Based on the dynamics model, a state-space equation incorporating the vehicle's motion state and disturbance input is constructed:

[0038] ;

[0039] in This is the vehicle state vector; It is a multi-source disturbance input, including external disturbances such as wind, waves, and currents, as well as internal disturbances such as dynamic system fluctuations; To control the input vector; To observe noise; This is the system matrix.

[0040] Set the performance index γ of the H∞ filter to ensure that the disturbance suppression capability is satisfied:

[0041] ;

[0042] in Let be the disturbance and state vector to be estimated. The estimated value is set as follows. Based on the typical scenario of sea state 3, the initial value of the core performance index γ of the H∞ filter is set to 1.2. The estimation result is directly transmitted to the subsequent optimized ADRC controller. The difference between the actual yaw angle and the expected yaw angle is calculated as the control error and compared with the preset 0.5 degree threshold. If the error is less than the threshold, it means that the current filtering effect meets the control requirements. If the error is greater than the threshold, the γ value is dynamically adjusted in steps of 0.1 until the error meets the requirements.

[0043] 2. Attitude control strategy based on optimized active disturbance rejection control

[0044] This invention employs a finite-time extended observer optimized active disturbance rejection control strategy, constructing an attitude control framework that includes AFTESO disturbance observation and attitude estimation, dynamic nonlinear control law, and actuator drive module. It completes accurate attitude estimation and disturbance compensation within a finite time, enabling rapid response, accurate tracking, and stable control of the amphibious heavy-duty transport vehicle's waterborne navigation attitude, thereby improving the vehicle's navigation safety and reliability under sudden disturbances.

[0045] As shown in Figure 4, the control system calculates the desired motion state of the vehicle using a vehicle dynamics model. Then, it uses an H∞ filter to initially filter out interference from system and environmental noise, obtaining the filtered desired state. The error between this filtered state and the actual motion state of the vehicle is calculated, reflecting the impact of external disturbances or system deviations on the vehicle. Finally, by optimizing the active disturbance rejection control strategy, the system generates the thruster's rotational speed based on the error, using it as input to the dynamics model and feeding it back to update the vehicle's forces and motion state, forming a complete closed-loop system.

[0046] As can be seen from the vehicle dynamics model, the system is a general second-order uncertain nonlinear system:

[0047] ;

[0048] in This is the system output (yaw angle). For control input (thruster speed); The known nonlinear dynamic equations of the system are... For coordinated interference; To control the gain.

[0049] Define state variables = , = , = The expected yaw angle and the system output yaw angle are the observation error. Then the dynamic equation of AFTESO is:

[0050] ;

[0051] in for Observations; Observer base gain; For adaptive gain; The adjustment parameter is for finite-time convergence.

[0052] Combining the ADRC control algorithm, the tracking differentiator extracts the tracking signal and differential signal of the reference input (yaw angle), while AFTESO estimates the system state and total disturbance in real time and uses the disturbance estimate directly for feedforward compensation of the control law, ensuring that the state error and disturbance estimation error of the entire closed-loop system can converge to zero within a finite time. The ADRC calculation yields the current thruster speed that can effectively suppress external and internal disturbances.

[0053] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0054] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0055] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An anti-interference control method for amphibious vehicles based on H∞ filtering and AFTESO optimized ADRC, characterized in that, The method includes: S1. Constructing a dynamic model of an amphibious heavy-duty transport vehicle based on vehicle parameters and real-time data collected by sensors, and solving for the desired lateral and longitudinal velocities and the desired yaw angle; S2. Performing noise suppression and preliminary disturbance identification on the collected thruster speed signals based on an H∞ filter to obtain the filtered desired state; S3. Based on the error between the filtered desired state and the actual motion state, using a finite-time extended state observer (AFTESO) to estimate the system state and total disturbance in real time; S4. Combining the active disturbance rejection control (ADRC) algorithm, using the total disturbance value estimated by AFTESO for feedforward compensation of the control law, calculating the thruster speed control command to suppress the disturbance, and feeding it back to the vehicle dynamic model to form a closed-loop control.

2. The method according to claim 1, characterized in that, In S1, the vehicle parameters include the vehicle's length, width, height, and body mass; the real-time data collected by the sensors includes the vehicle's draft, the spray pump motor speed, and driver input information.

3. The method according to claim 2, characterized in that, S1 also includes: calculating the thrust of the left and right propellers based on the external characteristic curve of the waterjet propeller, and calculating the thrust torque based on the rotating power arm; and calculating the yaw resistance torque based on the simplified engineering expression.

4. The method according to claim 3, characterized in that, The yaw resistance torque The calculation formula is: In the formula: The vehicle's angular velocity; The drag coefficient was subsequently obtained through parameter identification using real vehicle data. These are the actual width, length, and height of the vehicle; This is the density of water.

5. The method according to claim 3, characterized in that, The thrust of the left and right thrusters 、 The calculation formula is: ； ;in, The static thrust coefficient at a constant thruster speed. 、 The rotational speed of the thruster.

6. The method according to claim 1, characterized in that, S1 also includes: collecting data on angular velocity around the Z-axis, lateral and longitudinal vehicle speeds, and angular acceleration from actual vehicles, synchronizing the time and removing outliers, solving for the optimal parameters to be identified using the least squares method, and verifying the validity of the parameters through repeatability verification of multiple sets of data under the same working conditions.

7. The method according to claim 1, characterized in that, S2 specifically involves: using the rotational speed of the left and right thrusters as the control input, designing a discretized state-space model that covers external disturbances such as wind, waves, and currents, as well as internal dynamic structure disturbances; using an H∞ filter to perform noise suppression and preliminary disturbance identification on the rotational speed signals of the left and right thrusters collected in real time from the vehicle's rotational speed feedback; and dynamically adjusting the filtering performance index γ so that the difference between the actual yaw angle and the desired yaw angle meets a preset threshold.

8. The method according to claim 7, characterized in that, The state-space equation of the H∞ filter is: ;in This is the vehicle state vector; For multi-source disturbance input; To control the input vector; To observe noise; The system matrix is ​​given; the performance index γ of the H∞ filter satisfies: Among them Let be the disturbance and state vector to be estimated. Its estimated value.

9. The method according to claim 1, characterized in that, The system described by the dynamic model is a general second-order uncertain nonlinear system, and its expression is: ;in This is the system output, i.e., the yaw angle; The control input is the thruster speed; The known nonlinear dynamic equations of the system are... For coordinated interference; To control the gain, define state variables. = ， = ， = The expected yaw angle and the system output yaw angle are the observation errors. Then the dynamic equation of AFTESO is: in for Observations; Observer base gain; For adaptive gain; The adjustment parameter is for finite-time convergence.

Images