A control method for autonomous underwater vehicle considering saturation and dead zone characteristics of rudder

Abstract

The present invention relates to an autonomous underwater vehicle control method considering the saturation and dead zone characteristics of the rudder, and belongs to the field of automatic control technology. It includes: obtaining its own attitude and angular velocity, reference attitude angle; inputting the reference attitude angle into the command filter to obtain the reference trajectory and trajectory tracking error; designing the desired virtual control law, including an adaptive sliding mode controller, an adaptive neural network controller, and an adaptive dead zone inverse compensator, while considering the saturation characteristics of the rudder using a hyperbolic tangent function; solving the control input of the rudder based on a dynamic inverse method. The present invention processes the dead zone and saturation characteristics of the rudder, so that the original control input is converted into a corresponding ideal input value under saturation or dead zone, so that the controller has higher accuracy and can better exert the performance of the controller.

Description

Technical Field

[0001] The invention belongs to the technical field of automatic control and relates to nonlinear characteristic processing of a rudder during motion control of an underactuated autonomous underwater vehicle. Background Art

[0002] Autonomous Underwater Vehicle (AUV) has the advantages of small size, strong maneuverability, strong intelligence, high concealment and high operation safety. It can complete tasks such as ocean monitoring, marine archaeology and underwater operations without human intervention and large-scale surface support.

[0003] Compared with unmanned ships and surface vessels, the dynamic system of AUV is more complex, with nonlinear and time-varying characteristics. Due to the complex marine environment, AUV may be disturbed by ocean currents, surges, etc. during underwater navigation. In recent years, domestic and foreign experts have conducted relatively detailed research on motion control, committed to improving the maneuverability, stability, and robustness of AUV.

[0004] Most of the existing underactuated underwater vehicle control methods do not consider the nonlinear characteristics of the rudder, or only solve the saturation characteristics by direct limiting. This control method will cause damage to the rudder and mechanical structure in engineering practice. Some researchers have considered the rudder saturation characteristics or dead zone characteristics of underwater vehicle control, but have not considered both at the same time. Summary of the invention

[0005] Technical issues to be solved

[0006] In order to avoid the shortcomings of the prior art and take into account the saturation and dead zone characteristics of the rudder of the underwater vehicle so as to achieve higher control accuracy, better economy and safety, the present invention provides an autonomous underwater vehicle control method taking into account the saturation and dead zone characteristics of the rudder.

[0007] Technical Solution

[0008] A control method for an autonomous underwater vehicle considering the saturation and dead zone characteristics of a rudder, characterized by the following steps:

[0009] S1: Get its own attitude, angular velocity and reference attitude angle;

[0010] S2: Input the reference attitude angle into the command filter to obtain the reference trajectory and trajectory tracking error;

[0011] S3: Design the desired virtual control law, including an adaptive sliding mode controller, an adaptive neural network controller, and an adaptive dead zone inverse compensator, while taking into account the saturation characteristics of the rudder using the hyperbolic tangent function;

[0012] S4: Solve the rudder control input based on the dynamic inversion method.

[0013] Further technical solutions of the present invention: S1 is specifically: obtaining attitude information η2 and vehicle angular velocity v2 through the attitude sensor carried by the AUV itself; obtaining the reference heading angle ψ according to the task instructions issued by the host computer or the line of sight guidance method d With reference pitch angle θ d .

[0014] A further technical solution of the present invention: S2 is specifically: define the following command filter

[0015]

[0016] in, A d =diag{ζ r1 ω r1 ,ζ r2 ω r2 ,ζ r3 ω r3} is a positive definite parameter matrix, ω ri ,ζ ri , is the natural frequency and damping ratio of the filter; the reference trajectory is initialized by the initial state of the AUV, that is, η 2r (0) = η2(0) and v(0) is the speed of the AUV at time zero, η 2r is a smooth reference trajectory;

[0017] The trajectory tracking error is defined as

[0018]

[0019] Among them, e ψ e θ They are heading angle tracking error, pitch angle tracking error, and roll angle tracking error respectively;

[0020] The error after the first-order command filter is defined as

[0021]

[0022] Among them, Λ and K a is a positive constant gain vector, K a The value of depends on the relevant control force coefficient and the rudder limit; tanh is the hyperbolic tangent function, is the desired angular velocity as a virtual control of the attitude error.

[0023] A further technical solution of the present invention: S3 is specifically: designing the expected acceleration based on the approximate dynamic model, as shown in the following formula, so that the system approaches the sliding surface r = 0

[0024]

[0025] in, is the dead zone parameter estimation error, Related to the dead zone input, adaptive law is the transpose of the Jacobian matrix, d is the non-parameterizable dead zone estimation residual, is the estimated value of the upper bound of the residual, and is the positive definite parameter of the design;

[0026] u as is an adaptive sliding mode controller, defined as follows

[0027]

[0028] Among them, K s =diag{K s1 ,K s2 ,K s3} is a diagonal positive definite matrix, is the adaptive selection coefficient, K b is a positive constant gain matrix; the sign function sgn(r) is defined as

[0029]

[0030] Where r1, r2, r3 are the elements of the error r, sgn(·) is the scalar sign function;

[0031] The following neural network is designed to compensate for the uncertainty of the dynamics model

[0032]

[0033] in, is the estimation matrix of W and V;

[0034] Correspondingly, the estimation error is defined as u ar is a robust signal, designed as follows

[0035]

[0036] in, are two adaptive control parameters,

[0037] A further technical solution of the present invention is as follows: S4 is specifically: by using the pseudo-inverse method, the thrust control of the rudder δ is calculated using the approximate form of the dynamic model.

[0038]

[0039] in, yes The pseudo-inverse of , and this inverse must exist, because For an underactuated underwater vehicle, it is a full-rank matrix; yes The inverse function of d δ e δ r ] T , are differential rudder, horizontal rudder, vertical rudder respectively; and are the estimated system parameters and functions.

[0040] A computer system, characterized in that it comprises: one or more processors, and a computer-readable storage medium for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned method.

[0041] A computer-readable storage medium, characterized in that it stores computer-executable instructions, which are used to implement the above method when executed.

[0042] A computer program, characterized by comprising computer executable instructions, which are used to implement the above method when executed.

[0043] Beneficial Effects

[0044] The control method of an autonomous underwater vehicle considering the saturation and dead zone characteristics of the rudder proposed in the present invention has the following beneficial effects compared with the existing control methods:

[0045] 1. Strong robustness and high control accuracy

[0046] In the present invention, by processing the dead zone and saturation characteristics of the rudder, the original control input in saturation or dead zone is converted into a corresponding ideal input value, so that the controller has higher accuracy and can better exert the performance of the controller.

[0047] 2. Higher safety and better economy

[0048] In the present invention, by processing the saturation characteristics of the rudder, the efficiency and safety of the rudder are enhanced, and damage to the steering gear or mechanical parts such as the rudder plate and the connecting shaft caused by excessive input of the rudder is avoided. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The drawings are only for the purpose of illustrating particular embodiments and are not to be considered limiting of the present invention. Like reference symbols denote like components throughout the drawings.

[0050] Figure 1 Flow chart of the present invention;

[0051] Figure 2 AUV trajectory tracking trajectory;

[0052] Figure 3 X-axis tracking error and control input;

[0053] Figure 4 Y tracking error and control input;

[0054] Figure 5 Z tracking error and control input. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention 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 invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0056] The embodiment of the present invention provides an underwater vehicle control method based on an adaptive neural network and considering the saturation dead zone characteristics of the rudder. In the framework of adaptive control based on dynamic inversion, a sliding mode controller with neural network compensation is used to construct a virtual control signal κ, and a hyperbolic tangent function and an adaptive dead zone inverse compensator are used to process the nonlinear characteristics of the actuator to track a smooth reference trajectory η. 2r (t). The trajectory is determined by the target point position After passing through the command filter, the dynamic inverse control module converts the virtual signal into the actual propeller control signal through the approximate model of the underwater vehicle. Figure 1 As shown, the steps of the present invention are specifically as follows:

[0057] Step 1: Obtain attitude information η2 and vehicle angular velocity v2 through the attitude sensor carried by the AUV. Obtain the reference heading angle ψ according to the task instructions issued by the host computer or the line of sight guidance method d With reference pitch angle θ d , the reference roll angle is generally 0°.

[0058] Step 2: Define the following command filter

[0059]

[0060] in, A d =diag{ζ r1 ω r1 ,ζ r2 ω r2 ,ζ r3 ω r3} is a positive definite parameter matrix, ω ri ,ζ ri , is the natural frequency and damping ratio of the filter. The reference trajectory is initialized by the initial state of the AUV, that is, η 2r (0) = η2(0) and v(0) is the speed of the AUV at time zero, η 2r is a smooth reference trajectory.

[0061] The trajectory tracking error is defined as

[0062]

[0063] Among them, e ψ e θ They are heading angle tracking error, pitch angle tracking error, and roll angle tracking error respectively.

[0064] The error after the first-order command filter is defined as

[0065]

[0066] Among them, Λ and K a is a positive constant gain vector, K a The value of depends on the relevant control force coefficients and the rudder limits. tanh is the hyperbolic tangent function, is the desired angular velocity as a virtual control of the attitude error.

[0067] Step 3: Design the desired acceleration based on the approximate dynamic model, as shown in the following formula, so that the system approaches the sliding surface r = 0

[0068]

[0069] in, is the dead zone parameter estimation error, Related to the dead zone input, adaptive law is the transpose of the Jacobian matrix, d is the non-parameterizable dead zone estimation residual, is the estimated value of the upper bound of the residual, and is the positive definite parameter of the design.

[0070] u as is an adaptive sliding mode controller, defined as follows

[0071]

[0072] Among them, K s =diag{K s1 ,K s2 ,K s3} is a diagonal positive definite matrix, is the adaptive selection coefficient, K b is a positive constant gain matrix. The sign function sgn(r) is defined as

[0073]

[0074] where r1, r2, r3 are the elements of the error r and sgn(·) is the scalar sign function.

[0075] The following neural network is designed to compensate for the uncertainty of the dynamics model

[0076]

[0077] in, is the estimation matrix of W and V.

[0078] Correspondingly, the estimation error is defined as u ar is a robust signal, designed as follows

[0079]

[0080] in, are two adaptive control parameters,

[0081] Step 4: Calculate the thrust control of the rudder delta using the approximate form of the dynamic model by pseudo-inverse method

[0082]

[0083] in, yes The pseudo-inverse of , and this inverse must exist, because For an underactuated underwater vehicle, it is a full-rank matrix. yes The inverse function of δ=[δ d δ e δ r ] T , which are differential rudder, horizontal rudder and vertical rudder respectively. and are the estimated system parameters and functions.

[0084] In order to verify the effectiveness of the above control algorithm, the present invention also provides the following embodiments.

[0085] The initial state of the AUV is

[0086] x(0)=-3 y(0)=1 z(0)=-3 φ(0)=0 θ(0)=0 ψ(0)=0

[0087] u(0)=0.5 v(0)=0.5 w(0)=0 p(0)=0 q(0)=0 r(0)=0

[0088] The initial position of the target is (0.25, 0, 0), and it moves along the x-axis at a uniform speed of 0.1 m / s. Set the dynamic target tracking task so that the final desired state of the AUV is

[0089] Δx=-2,Δy=0,Δz=0, Δθ=0,Δψ=0

[0090] Tracking results such as Figure 2-5 shown.

[0091] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should be included in the protection scope of the present invention.

Claims

1. A control method for an autonomous underwater vehicle considering the saturation and dead zone characteristics of a rudder, characterized in that Here are the steps: S1: Get its own attitude, angular velocity and reference attitude angle; S2: Input the reference attitude angle into the command filter to obtain the reference trajectory and trajectory tracking error; S3: Design the desired virtual control law, including adaptive sliding mode controller, adaptive neural network controller, and adaptive dead zone inverse compensator, and use the hyperbolic tangent function to consider the saturation characteristics of the rudder; specifically: design the desired acceleration based on the approximate dynamic model, as shown in the following formula, so that the system approaches the sliding mode surface r = 0 in, is the dead zone parameter estimation error, Related to the dead zone input, adaptive law is the transpose of the Jacobian matrix, d is the non-parameterizable dead zone estimation residual, is the estimated value of the upper bound of the residual, and is the positive definite parameter of the design; u as is an adaptive sliding mode controller, defined as follows Among them, K s =diag{K s1 ,K s2 ,K s3 } is a diagonal positive definite matrix, is the adaptive selection coefficient, K b is a positive constant gain matrix; the sign function sgn(r) is defined as Where r1, r2, r3 are the elements of the error r, sgn(·) is the scalar sign function; The following neural network is designed to compensate for the uncertainty of the dynamics model in, is the estimation matrix of W and V; Correspondingly, the estimation error is defined as u ar is a robust signal, designed as follows in, are two adaptive control parameters, S4: Solve the rudder control input based on the dynamic inversion method.

2. The autonomous underwater vehicle control method considering the saturation and dead zone characteristics of the rudder according to claim 1 is characterized in that S1 is specifically as follows: obtain the attitude information η2 and the angular velocity v2 of the vehicle through the attitude sensor carried by the AUV itself; obtain the reference heading angle ψ according to the task instructions issued by the host computer or the line of sight guidance method d With reference pitch angle θ d .

3. The autonomous underwater vehicle control method considering the saturation and dead zone characteristics of the rudder according to claim 2, characterized in that S2 is specifically defined as follows: in, A d =diag{ζ r1 ω r1 ,ζ r2 ω r2 ,ζ r3 ω r3 } is a positive definite parameter matrix, is the natural frequency and damping ratio of the filter; the reference trajectory is initialized by the initial state of the AUV, that is, η 2r (0) = η2(0) and v(0) is the speed of the AUV at time zero, η 2r is a smooth reference trajectory; The trajectory tracking error is defined as in, They are heading angle tracking error, pitch angle tracking error, and roll angle tracking error respectively; The error after the first-order command filter is defined as Among them, Λ and K a is a positive constant gain vector, K a The value of depends on the relevant control force coefficient and the rudder limit; tanh is the hyperbolic tangent function, is the desired angular velocity as a virtual control of the attitude error.

4. The autonomous underwater vehicle control method considering the saturation and dead zone characteristics of the rudder according to claim 1, characterized in that S4 is specifically: by using the pseudo-inverse method, the thrust control of the rudder δ is calculated using the approximate form of the dynamic model in, yes The pseudo-inverse of , and this inverse must exist, because For an underactuated underwater vehicle, it is a full-rank matrix; yes The inverse function of d δ e δ r ] T , are differential rudder, horizontal rudder, vertical rudder respectively; and are the estimated system parameters and functions.

5. A computer system, characterized in that include: One or more processors, and a computer-readable storage medium for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the method of claim 1.

6. A computer-readable storage medium, characterized in that Computer executable instructions are stored, and when the instructions are executed, they are used to implement the method of claim 1.

7. A computer program product, characterized in that The invention comprises computer executable instructions, which are used to implement the method of claim 1 when being executed.

Images