Model-free adaptive control method for depth tracking of underactuated unmanned underwater vehicle

Abstract

The present application discloses a model-free adaptive control method for depth tracking of underactuated unmanned submersibles, belonging to the field of unmanned submersible control technology. First, the depth error between the actual depth and the expected depth of the submersible is converted into a pitch error, and the expected pitch angle with angle of attack compensation is obtained based on the adaptive line-of-sight guidance law; then the expected pitch angle is adjusted using a tracking differentiator; finally, based on a dynamic linearization data model, the depth error, pitch error, expected pitch angle and tracking differentiator are combined to obtain the expected control rudder angle of the submersible. The present application can estimate and compensate for the angle of attack and dynamic linearization residual in real time, further improving the comprehensive tracking performance. When faced with the influence of external sudden disturbances, it has faster convergence speed, higher steady-state accuracy and stronger anti-interference ability compared with classical control methods. At the same time, the steering is smooth and changes slowly, which can significantly reduce mechanical noise, reduce fatigue damage of the servo and extend service life.

Description

Technical Field

[0001] The present application belongs to the field of unmanned underwater vehicle control technology, and more specifically, relates to a model-free adaptive control method for depth tracking of an underactuated unmanned underwater vehicle. Background Art

[0002] In recent years, the application of autonomous underwater vehicles (UUVs) in underwater exploration has attracted widespread attention. Compared with manned submersibles and remotely operated vehicles (ROVs), UUVs have more advantages, including flexibility, autonomy, and size diversity. These characteristics make them more suitable for underwater operations, including but not limited to terrain scanning, marine resource exploration, pipeline inspection, underwater search and rescue, etc. In order to ensure that UUVs can perform accurate and efficient performance in three-dimensional space motion, dynamic models and precise control algorithms play a vital role. In addition, the motion control and depth tracking control technology of UUVs are of great significance for achieving efficient motion, and accurate depth tracking control technology plays an important role in performing underwater operations. Due to the nonlinearity, strong coupling and parameter uncertainty of underwater unmanned submersibles, these characteristics make their accurate path tracking control face great challenges. The model-free control method provides a reliable and effective solution to this problem by replacing the traditional system mathematical model with a model based on I / O data obtained by dynamic linearization. However, this method inevitably excludes the high-order terms of the model, resulting in the existence of unmodeled dynamics, which will further reduce the comprehensive control performance after superimposing external disturbances. Summary of the invention

[0003] In response to the above defects or improvement needs of the prior art, the present application provides a model-free adaptive control method for depth tracking of an underactuated unmanned underwater vehicle, which aims to solve the technical problem that the control performance of the existing model-free control method is reduced after being applied to depth tracking of an underactuated unmanned underwater vehicle.

[0004] To achieve the above objectives, in a first aspect, the present application provides a model-free adaptive control method for depth tracking of an underactuated unmanned underwater vehicle, comprising:

[0005] The depth error between the actual depth and the expected depth of the submersible is converted into a pitch error, and the expected pitch angle with angle of attack compensation is obtained based on the adaptive line-of-sight guidance law;

[0006] The desired trim angle is adjusted using a tracking differentiator;

[0007] Based on the dynamic linearization data model, the desired control rudder angle of the submersible is obtained by combining the depth error, the pitch error, the desired pitch angle and the tracking differentiator.

[0008] Preferably, the depth error between the actual depth and the expected depth of the submersible is converted into a pitch error, specifically:

[0009]

[0010]

[0011]

[0012] Among them, e x and e z is the axial error and lateral error of the tracking point coordinate system; (x i , z i ) is the position of the i-th waypoint on the vertical path, (x i+1 , z i+1 ) is the position of the i+1th waypoint on the vertical path, is the rotation matrix, γ path is the path tangent angle of the vertical plane, and its value range is (-π / 2, π / 2).

[0013] Preferably, the adaptive line-of-sight guidance law is:

[0014]

[0015]

[0016] Among them, θ d is the desired pitch angle, γ path is the path tangent angle of the vertical plane, is the estimated angle of attack, for The differential form of e z is the lateral error of the tracking point coordinate system, Δ is the forward distance along the x-axis in the tracking point coordinate system, and k is the adaptive gain.

[0017] Preferably, the dynamic linearization data model is:

[0018] Δθ(s+1)=Φ p (s)Δδ s (s)+ξ(s)

[0019] Among them, Δθ(s+1) is the change in the pitch angle at time s+1, Φ p (s)∈R represents the pseudo partial derivative, which is obtained by the pseudo partial derivative function estimation algorithm, R is a real number set, Δδ s (s) is the change of the rudder angle input at time s, and ξ(s) is the total change of the unmodeled and external disturbances of the linearized data model at time s.

[0020] Preferably, the desired control rudder angle is:

[0021]

[0022] Among them, δ s (s) and δ s (s-1) are the desired control rudder angles at time s and time s-1, ρ∈(0,1] is the step size parameter, λ>0 is the weight parameter for limiting the change of system input, and k p and k d To control the gain, is the total change estimate of the unmodeled and external disturbances of the linearized data model at time s, Φ p (s)∈R represents the pseudo partial derivative, which is obtained by the pseudo partial derivative function estimation algorithm, and R is a real number set; e θ (s) and e q (s) is the tracking error:

[0023] e θ (s) = x1(s+1)-θ(s)

[0024] e q (s) = x2(s+1)-q(s)

[0025] x1(s+1) is the expected pitch angle at time s+1 after adjustment by the tracking differentiator, x2(s+1) is the differential signal of the tracking differentiator tracking the pitch angle at time s+1, θ(s) and q(s) are the actual pitch angle and actual pitch angular velocity of the submersible at time s, respectively.

[0026] Preferably, the pseudo partial derivative function estimation algorithm is specifically:

[0027]

[0028] in, and is the pseudo partial derivative Φ at time s and time s-1 p (s) estimated value, η∈(0,1] is the introduced step size factor, μ is the penalty factor for the change of the estimated value of the pseudo partial derivative function, Δδ s (s-1) is the change in the desired control rudder angle at time s-1, is the total change estimate of the unmodeled and external disturbances of the linearized data model at time s-1, ΔY(s) is the state reorganization change, ΔY(s) = k p Δθ(s)+k d Δq(s),k p and k d is the control gain, Δθ(s) and Δq(s) are the actual change in pitch angle and actual change in pitch angular velocity of the submersible at time s, respectively.

[0029] If satisfied |Δδ s (s-1)|≤ε or but Among them, ε is a preset constant and sign() is a sign function.

[0030] Preferably, the total change of the unmodeled and external disturbances of the linearized data model is observed based on a discrete-time extended state observer, and the discrete-time extended state observer is:

[0031]

[0032]

[0033] Among them, ξ(s) and ξ(s+1) are the total changes of the unmodeled and external disturbances of the linearized data model at time s and time s+1, z1(s)=θ(s), z2(s)=ξ(s), θ(s) is the actual pitch angle of the submersible at time s, and are the estimated values ​​of z1(s) and z2(s), Φ p (s) is the pseudo partial derivative, is Φ p The estimated value of (s), Δδ s (s) is the change in the rudder angle input at time s, h is the sampling step length, β 01 and β 02 is the observer gain, e1(s) and e2(s) are the estimation errors,

[0034] In a second aspect, the present application provides an electronic device, comprising: a memory for storing programs; a processor for executing the programs stored in the memory, when the program stored in the memory is executed, the processor is used to execute the method described in the first aspect or any possible implementation of the first aspect.

[0035] In a third aspect, the present application provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the method described in the first aspect or any possible implementation of the first aspect.

[0036] In a fourth aspect, the present application provides a computer program product. When the computer program product runs on a processor, the processor executes the method described in the first aspect or any possible implementation of the first aspect.

[0037] In general, the above technical solutions conceived by this application have the following beneficial effects compared with the prior art:

[0038] (1) The present application provides a model-free adaptive control method for depth tracking of underactuated underwater unmanned submersibles. In view of the nonlinear, strong coupling and uncertainty characteristics of underactuated submersible systems, an anti-interference, adaptive model-free two-layer tracking control method from the outside to the inside is designed. This method solves the problem of less control and more control of underactuated systems by converting pitch errors into depth errors. It estimates and adaptively compensates for the angle of attack generated by the vertical velocity in real time, overcomes the guidance deviation introduced by residual buoyancy or environmental disturbances, and improves the guidance accuracy from the outer loop kinematic level. At the same time, the present application designs a tracking differentiator for the expected input pitch angle, smoothly transitions the tracking input signal, and avoids excessive changes in the expected input pitch angle, which would cause unreasonable phenomena.

[0039] (2) The inner-loop dynamics layer of this application adopts an incremental IMFAC control algorithm with high-order unmodeled dynamics and external disturbance compensation, which significantly reduces the output rudder angle and frequency, and can effectively improve energy efficiency, reduce mechanical noise and extend the service life of the actuator. The entire control method does not need to be based on a system model, and is particularly suitable for submersibles with complex modeling or low precision. A discrete-time extended state observer (DESO) was constructed to effectively predict the residual nonlinear internal and external disturbances and uncertainties generated by the dynamic linearization data modeling process, thereby improving the anti-interference ability and comprehensive control performance of underwater unmanned submersibles.

[0040] (3) In order to enable the traditional IMFAC method to be used on a rotatable unmanned underwater vehicle, this application designs a method that combines the state error and the differential signal of the error into a linear state error feedback law (LSEF), which can effectively reduce delay and overshoot, reduce oscillation and shorten convergence time, and further improve tracking performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a schematic diagram of a model-free adaptive control method provided in an embodiment of the present application;

[0042] Figure 2 It is a schematic diagram of parameter definition of an underactuated underwater unmanned submersible system provided in an embodiment of the present application;

[0043] Figure 3 is a schematic diagram of a model-free adaptive control process provided by an embodiment of the present application;

[0044] Figure 4 This is a diagram of a test scenario of an underactuated underwater unmanned submersible provided in an embodiment of the present application;

[0045] Figure 5 is an attack angle estimation curve diagram based on an adaptive line-of-sight guidance law provided in an embodiment of the present application;

[0046] Figure 6It is an improved pseudo partial derivative estimation diagram of an underactuated underwater unmanned submersible provided in an embodiment of the present application;

[0047] Figure 7 It is a residual estimation diagram of the dynamic linearization model of the underactuated underwater unmanned submersible provided in the embodiment of the present application;

[0048] Figure 8 is a curve diagram of the elevator rudder angle of an underactuated underwater unmanned submersible provided in an embodiment of the present application;

[0049] Fig. 9 It is a depth tracking curve diagram of an underactuated underwater unmanned submersible provided in an embodiment of the present application;

[0050] Fig.10 It is a tracking curve diagram of the pitch state of an under-actuated underwater unmanned submersible provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0052] First, in view of the underactuated characteristics of underwater unmanned submersibles, a strategy of controlling more with less is adopted, and the control parameters need to be converted to solve the underactuated control problem. At the same time, facing the influence of residual buoyancy and external environmental disturbances, it is necessary to design more advanced guidance theories and methods, such as ILOS, ALOS, ELOS, etc., so as to effectively compensate for the angle of attack problem that is easily ignored. For the dynamic linearization modeling process, it is necessary to design a more accurate model error observer to compensate for the residual dynamic terms and external disturbances that are not modeled in the dynamic linearization process.

[0053] The purpose of this application is to first transform the depth tracking problem into a pitch tracking problem by using an adaptive line-of-sight LOS guidance law, solve the underactuated control problem, and compensate for the angle of attack disturbance that is usually ignored in the classical LOS algorithm. Secondly, the expected input pitch angle in the MFAC control algorithm is optimized, and a tracking differentiator is used to ensure a smooth transition of the pitch state tracking. In addition, a discrete-time extended state observer (DESO) is used to estimate and compensate for the dynamic and environmental disturbances that are not modeled in the dynamic linearization data model. Finally, in order to verify the effectiveness and superior control performance of the proposed method, a tank test was carried out on a lightweight underactuated underwater unmanned submersible to ensure the rationality and practicality of the proposed method.

[0054] Figure 1 A summary diagram of the model-free adaptive (ALOS-IMFAC) control method for depth tracking of an underactuated underwater unmanned submersible provided in an embodiment of the present application; Figure 1 As shown, it includes the following components:

[0055] S101, uses adaptive line-of-sight guidance with angle-of-attack compensation to convert depth error into pitch error, solving the underactuated control problem;

[0056] S102, using a tracking differentiator to adjust the required input parameters and improve transient performance;

[0057] S103, designed an extended state observer in discrete time to predict and compensate for high-order unmodeled dynamics and external disturbances

[0058] S104, using an I / O data-driven model-free adaptive (IMFAC) controller with double-layer compensation, solving an approximate digitized model based on a dynamic linearization method, and combining the depth plane error and the pitch angle error, the line of sight guidance law derives the desired pitch angle, the tracking differentiator arranges the transition process, and the discrete expanded state observer compensates for unmodeled dynamics and disturbances.

[0059] Figure 2 A schematic diagram of the definition of parameters of an underactuated underwater unmanned submersible system provided in an embodiment of the present application; Figure 2 As shown, the basic parameters of the underactuated underwater unmanned submersible system used in this embodiment are shown in Table 1:

[0060] Table 1

[0061]

[0062] Figure 3 A schematic diagram of a model-free adaptive ALOS-IMFAC control flow for depth tracking of an underactuated underwater unmanned submersible provided in an embodiment of the present application; Figure 3 In, z d is the desired tracking depth, z(s) is the real-time measured depth, θ d (s) is the guidance pitch angle, θ(s) is the real-time measured pitch angle, q(s) is the real-time measured pitch angular velocity. x1(s) is the required pitch angle after the tracking differentiator transition arrangement, and x2(s) is the differential signal after the tracking differentiator transition arrangement.

[0063] Specifically, this application designs a dual-loop model-free adaptive control method for an underactuated underwater unmanned submersible:

[0064] Firstly, an adaptive line-of-sight guidance law is adopted in the outer loop kinematic guidance layer to convert the depth error into the pitch error to solve the control problem of the underactuated system. At the same time, the angle of attack is estimated and compensated in real time to solve the influence of its own residual buoyancy and external environmental disturbance on the guidance accuracy.

[0065] Secondly, a tracking differentiator is designed to smooth the expected input tracking signal and arrange the differential signal to solve the problem of overshoot and oscillation caused by too fast adjustment;

[0066] Finally, a model-free adaptive controller with dynamic residual model error compensation is designed in the inner loop dynamic control layer. At the same time, a linear state error feedback law is designed to combine the input and measurement states to improve the transient performance and steady-state accuracy of the underactuated underwater unmanned submersible. The dynamic linearization modeling residual and environmental disturbances are compensated in real time to comprehensively improve the anti-disturbance ability of the submersible from the outer layer to the inner layer. The controller is:

[0067]

[0068] Among them, δ s (s) and δ s (s-1) represents the steering angle input at the current moment and the previous moment respectively, ρ∈(0,1] is the step size parameter, λ>0 is the weight parameter to limit the change of system input, k p and k d is the control gain that needs to be adjusted, is the total change of the unmodeled terms and external disturbances estimated by the discrete-time extended state observer DESO, e θ (s) = θ d (s+1)-θ(s)=x1(s+1)-θ(s),e q (s) = q d (s+1)-q(s)=x2(s+1)-q(s) is the tracking error after smoothing, x1(s+1) is the expected pitch angle after the tracking differentiator TD transition arrangement, and x2(s+1) is the differential signal of the pitch angle of TD tracking. Figure 3 As can be seen from the flow chart, the input of the tracking differentiator is the output of the adaptive line-of-sight guidance law.

[0069] For example, the adaptive line-of-sight guidance law is:

[0070]

[0071]

[0072] Among them, θ d is the desired pitch angle, γ path is the path tangent angle of the vertical plane, is the estimated value of the unknown angle of attack α, is a differential form, e z is the lateral error of the tracking point coordinate system, Δ is the forward distance along the x-axis in the tracking point coordinate system, and k is the adaptive gain that needs to be adjusted.

[0073] For example, the DESO observer (discrete time extended state observer) is:

[0074]

[0075]

[0076] Where ξ(s) is the total disturbance in the form of unknown increments, z1(s) = θ(s) and z2(s) = ξ(s), and are the estimated values ​​of z1(s) and z2(s), Φ p (s) is the pseudo partial derivative of the dynamic linearization solution, which can be given by the pseudo partial derivative function estimation algorithm, h is the system sampling step, β 01 , β 02 is the observer gain, is the estimation error.

[0077] For example, the pseudo partial derivative function estimation algorithm is:

[0078]

[0079] if |Δδ s (s-1)|≤ε or So

[0080]

[0081] Among them, η∈(0,1] is the introduced step size factor, μ is the penalty factor for the change of the estimated value of the pseudo partial derivative function. ε represents a sufficiently small positive constant, is the pseudo partial derivative function Φ p The estimated value of (s), Indicates reset to initial value. ΔY(s)=k p Δθ(s)+k d Δq(s) is the change in state reorganization, which is calculated by the linear state difference law. The estimated value of the total change ξ(s) of the residual unmodeled part of the dynamic linearized data model and the external disturbance that needs to be compensated is obtained by solving the observer DESO.

[0082] For example, the dynamic linearization data model is:

[0083] Δθ(s+1)=Φ p (s)Δδ s (s)+ξ(s)

[0084] Among them, Δθ(s+1) is the change in pitch angle, Φ p(s)∈R represents the pseudo partial derivative, which can be obtained by the pseudo partial derivative function estimation algorithm, Δδ s (s) is the change of the rudder angle input, and ξ(s) is the total change of the residual unmodeled and external disturbance of the linearized data model.

[0085] The tracking differentiator is:

[0086]

[0087] Among them, x1(s+1) is the expected pitch angle after the tracking differentiator TD transition arrangement, and x2(s+1) is the differential signal of the pitch angle tracked by TD. v(s) is the input signal, m0 is the integration step, r is the tracking speed factor, and fun is the discrete system fastest control comprehensive function. The specific expression is:

[0088]

[0089] The linear state error feedback law is:

[0090] E Y (s)=Y d -Y(s)=k p e θ (s)+k d e q (s)

[0091] Among them, Y d is the expected combination state, Y(s)=k p θ(s)+k d q(s) is the measurement combination state, k p and k d is the control gain that needs to be adjusted.

[0092] The linear state difference law is:

[0093] ΔY(s)=k p Δθ(s)+k d Δq(s)

[0094] Among them, Δθ(s) is the measured pitch state change, Δq(s) is the measured state pitch angular velocity change, k p and k d The control gain that needs to be adjusted

[0095] It is understood that the parameters required in all the calculation processes involved in this application can be determined by those skilled in the art based on experience or simulation experiments, so as to obtain the best control effect; this application does not impose any restrictions on the specific parameter values. Among them, the control effect refers to the effects of depth tracking, pitch angle tracking, etc., that is, the tracking error between the real-time measured depth and the expected depth and the tracking error between the real-time measured pitch angle and the expected pitch angle.

[0096] In a specific embodiment, this embodiment uses a towing tank test to verify the effectiveness and superiority of the method proposed in this application. The scenario of an underwater unmanned submersible conducting a deep tracking test in a tank is as follows: Figure 4 As shown. In order to verify the effectiveness and superiority of the method proposed in this application, the classic PID controller is used as a comparative verification method for this experiment. During the fixed-depth navigation process, in order to simulate the underwater unmanned submersible being disturbed by uncertainty, the unmanned submersible's onboard suction and drainage system is used to continuously suck water into the watertight compartment for 5 seconds from the beginning of the dive to 60 seconds, thereby increasing the mass of the submersible and reducing the residual buoyancy. By plotting the stored test data, Figure 5 The angle of attack of the unmanned underwater vehicle estimated by the adaptive line of sight guidance algorithm is given. It should be noted that the movement is too violent at the beginning of the dive, and the angle of attack estimation is inaccurate. The lack of angle of attack compensation at the beginning does not affect the subsequent steady-state accuracy. As it slowly approaches the expected depth, the diving speed gradually slows down, and the angle of attack estimation gradually stabilizes. Figure 6 and Figure 7 The estimation of the pseudo partial derivative of the dynamic linearization data model and its residual error are given respectively. The pseudo partial derivative is a bounded slowly time-varying parameter, does not contain any model information of the controlled system, is insensitive to time-varying parameters and time-varying structures, and has strong adaptability and robustness.

[0097] Test results Figure 8 A comparison of the elevator angle actions using the PID control algorithm and the method of the present invention is given. It can be clearly seen that the PID algorithm operates a sudden saturation rudder at the beginning of the dive, and then tends to be stable. During the stabilization process, the rudder angle still changes frequently, and there is a chattering phenomenon. The steering action of the method of the present invention is relatively slow, there is no sudden change in the initial steering, and the rudder angle changes smoothly after stabilization, which helps to reduce mechanical noise, save energy consumption, and at the same time reduce the steering frequency to extend the service life of the servo.

[0098] Test results Fig. 9 A comparison of the depth tracking effects of the PID control algorithm and the method of the present invention is given. Fig.10 The expected states of the two methods are compared. Since the method of the present invention tracks the combined state of the pitch angle and the pitch angular velocity, the existence of the pitch angular velocity causes some chattering in the expected reorganized state. Fig. 9It can be seen that the PID method takes about 50 seconds to converge to the desired depth of 2 meters, and the method of the present invention takes about 40 seconds. Compared with the PID control method, the present method can converge to the desired depth faster than the PID. At about the 60th second of sailing, the suction and drainage system began to absorb water into the watertight compartment to increase the mass of the hull. After being disturbed by water absorption, both methods were affected and deviated from the stable value and sank downward. The difference in the settlement depth of the two methods was not much, about 0.08 meters, and both were able to resist disturbances and recover to a stable state. However, it can be clearly seen that the method of the present invention has a faster recovery speed after being disturbed by water absorption, and the steady-state accuracy after recovery is higher.

[0099] In summary, it can be seen that the ALOS-IMFAC control scheme proposed by the method of the present invention can estimate and compensate for the angle of attack and dynamic linearization residual in real time, further improve the comprehensive tracking performance, and has a faster convergence speed and higher steady-state accuracy compared with the classical PID method. When facing the influence of external sudden disturbances, the proposed tracking control scheme can still show excellent anti-disturbance ability, recovery speed and steady-state accuracy. At the same time, the steering is smooth and changes slowly, which can significantly reduce mechanical noise, reduce servo fatigue damage and extend service life.

[0100] The present application discloses a model-free adaptive ALOS-IMFAC control method for deep tracking of underactuated unmanned underwater vehicles. The proposed method includes a two-layer architecture, including an outer loop kinematic guidance layer and an inner loop dynamic control layer, and has dual interference suppression functions of guidance compensation and residual compensation. In the outer loop kinematic layer, an adaptive line-of-sight guidance law ALOS with angle of attack compensation is used to convert the depth error into a pitch error to solve the underactuated control problem. In addition, a tracking differentiator is used to adjust the required input parameters and improve transient performance. In the inner loop dynamic layer, an improved model-free adaptive IMFAC tracking algorithm based on an I / O data-driven method is designed. The mathematical model of the nonlinear system is dynamically linearized into an equivalent data model, which does not contain any model information of the controlled system and is a data-driven controller. In the proposed IMFAC tracking scheme, an extended state observer in discrete time form is combined to predict and compensate for high-order unmodeled dynamics and external disturbances, thereby improving the accuracy of the dynamic linearization data model. The proposed scheme has been transplanted into the embedded system of underwater unmanned submersibles, and the fixed depth tracking test has been carried out in the towing tank, and the experimental comparison results have been obtained. These results verify the effectiveness, feasibility and superiority of the proposed scheme in realizing the depth tracking control of underwater unmanned submersibles, especially for nonlinear systems with complex modeling, low modeling accuracy or difficult modeling.

[0101] Based on the method in the above embodiment, an embodiment of the present application provides an electronic device, which may include: a processor (processor) 810, a communication interface (Communications Interface) 820, a memory (memory) 830 and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other through the communication bus 840. The processor 810 can call the logic instructions in the memory 830 to execute the method in the above embodiment.

[0102] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application.

[0103] Based on the method in the above embodiment, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the method in the above embodiment.

[0104] Based on the method in the above embodiment, an embodiment of the present application provides a computer program product. When the computer program product runs on a processor, the processor executes the method in the above embodiment.

[0105] It is understood that the processor in the embodiments of the present application may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0106] The method steps in the embodiments of the present application can be implemented by hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, and the software modules can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, mobile hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC.

[0107] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a storage medium or transmitted through the storage medium. The computer instructions may be transmitted from a website site, a computer, a server or a data center to another website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The storage medium may be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.

[0108] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0109] It should be understood that the various numerical numbers involved in the embodiments of the present application are only used for the convenience of description and are not used to limit the scope of the embodiments of the present application.

[0110] The above content is easily understood by those skilled in the art. The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A model-free adaptive control method for depth tracking of an underactuated unmanned underwater vehicle, characterized in that: include: The depth error between the actual depth and the expected depth of the submersible is converted into a pitch error, and the expected pitch angle with angle of attack compensation is obtained based on the adaptive line-of-sight guidance law; The desired trim angle is adjusted using a tracking differentiator; Based on the dynamic linearization data model, the desired control rudder angle of the submersible is obtained by combining the depth error, the pitch error, the desired pitch angle and the tracking differentiator; The desired control rudder angle is: in, and They are Moment and The expectation of the moment controls the rudder angle, is the step size parameter, To limit the weight parameter of the system input variation, and To control the gain, for An estimate of the total change in the unmodeled and external disturbances of the linearized data model at time instants, represents the pseudo partial derivative, which is obtained by the pseudo partial derivative function estimation algorithm. is the set of real numbers; and is the tracking error: After adjustment by the tracking differentiator Always expect the trim angle, To track the differentiator The differential signal of the pitch angle at the moment, and Submersible The actual pitch angle and actual pitch angular velocity at the moment.

2. The model-free adaptive control method according to claim 1, characterized in that: The depth error between the actual depth of the submersible and the expected depth is converted into the pitch error, specifically: in, and is the axial error and lateral error of the tracking point coordinate system; The vertical path The location of the waypoints, The vertical path The location of the waypoints, is the rotation matrix, is the path tangent angle of the vertical plane, and its value range is .

3. The model-free adaptive control method according to claim 1, characterized in that: The adaptive line-of-sight guidance law is: in, is the desired pitch angle, is the path tangent angle of the vertical plane, is the estimated angle of attack, for The differential form of is the lateral error of the tracking point coordinate system, is the foresight distance along the x-axis in the tracking point coordinate system, is the adaptive gain.

4. The model-free adaptive control method according to claim 1, characterized in that: The dynamic linearization data model is: in, for The change of pitch angle at a moment, represents the pseudo partial derivative, which is obtained by the pseudo partial derivative function estimation algorithm. is the set of real numbers, for The change of rudder angle input at each moment, for The total change in the unmodeled and external disturbances of the linearized data model at time instant.

5. The model-free adaptive control method according to claim 1 or 4, characterized in that: The pseudo partial derivative function estimation algorithm is specifically as follows: in, and for Moment and Pseudo-partial derivative at time Estimated value, is the introduced step size factor, is the penalty factor for the change in the estimated value of the pseudo partial derivative function, for The desired control of the rudder angle change at each moment, for An estimate of the total change in the unmodeled and external disturbances of the linearized data model at time instants, is the state reorganization change, , and To control the gain, and Submersible The actual change in pitch angle and the actual change in pitch angular velocity at the moment; If satisfied , or ,but ,in, is the preset constant, is a sign function.

6. The model-free adaptive control method according to claim 1 or 4, characterized in that: Based on the discrete time extended state observer, the total change of the unmodeled and external disturbance of the linearized data model is observed. The discrete time extended state observer is: in, and for Moment and The total change of the unmodeled and external disturbances of the linearized data model at time instant, , , For submersible The actual trim angle at the moment, and They are and The estimated value of is the pseudo partial derivative, for The estimated value of for The change of rudder angle input at each moment, is the sampling step length, and is the observer gain, and is the estimation error, .

7. An electronic device, characterized in that: include: Memory for storing computer programs; A processor, configured to execute the program stored in the memory. When the program stored in the memory is executed, the processor is configured to execute the method according to any one of claims 1 to 6.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program runs on a processor, the processor is caused to execute the method according to any one of claims 1 to 6.

9. A computer program product, characterized in that When the computer program product runs on a processor, the processor is caused to execute the method according to any one of claims 1 to 6.

Images