Deep sea hydraulic mechanical arm super-spiral sliding mode control method based on observer

Abstract

The invention discloses an observer-based super-spiral sliding mode control method for a deep-sea hydraulic mechanical arm, which comprises the following steps of: firstly, establishing a nonlinear dynamic model of the deep-sea hydraulic mechanical arm, and providing a finite time convergence observer according to the established nonlinear dynamic model to estimate the lumped disturbance of a deep-sea hydraulic mechanical arm system; and feedforward compensation is carried out in a subsequent controller design process. According to the method, the self-adaptive law design and the finite time convergence observer are fused, the super-spiral sliding mode controller is designed, the valve port control law of the deep-sea hydraulic mechanical arm servo valve is obtained, accurate control over the deep-sea hydraulic mechanical arm is achieved, and the problem that an existing control method is insufficient in control precision is solved.

Description

Technical Field

[0001] The present invention belongs to the field of motion control of deep - sea operation manipulators, and relates to a super - twisting sliding - mode control method, specifically to a super - twisting sliding - mode control method for a deep - sea hydraulic manipulator based on a finite - time convergent observer. Background Art

[0002] With the gradual deepening of human exploration and understanding of the ocean, deep - sea scientific research, resource exploration and operation tasks are increasing continuously, and the demand for deep - sea operation equipment is becoming increasingly urgent. The deep - sea hydraulic manipulator, as an adaptable operation device for equipment such as deep - sea submersibles and robots, plays an important role in tasks such as in - situ deep - sea operation, cable laying, and seabed rescue. However, achieving precise joint position control is a necessary prerequisite for completing the above - mentioned operation tasks. At present, PID control, as a common control method for deep - sea hydraulic manipulators, is difficult to achieve precise control when facing the highly nonlinear and strongly uncertain deep - sea hydraulic manipulator system. In addition, during the deep - sea operation process of the deep - sea hydraulic manipulator, it usually encounters external disturbances such as undersea undercurrents. These influencing factors will affect the precise control of the deep - sea hydraulic manipulator and also pose more stringent requirements on the robustness and anti - interference ability of the deep - sea hydraulic manipulator control method. Therefore, in order to further improve the operation ability of the deep - sea hydraulic manipulator, it is urgent to explore high - performance controllers to achieve precise control of the deep - sea hydraulic manipulator to ensure the efficient completion of seabed operation tasks. Summary of the Invention

[0003] Aiming at the problems of high nonlinearity, strong uncertainty and external disturbances faced by the deep - sea hydraulic manipulator during the movement process, the present invention proposes an observer - based super - twisting sliding - mode control method for the deep - sea hydraulic manipulator. This method proposes a finite - time convergent observer to estimate the lumped disturbance of the deep - sea hydraulic manipulator system through the established nonlinear dynamic model. By integrating the engineering mapping adaptive law and the finite - time convergent observer, a super - twisting sliding - mode controller is designed to achieve precise control of the deep - sea hydraulic manipulator and solve the problem of insufficient control accuracy existing in the existing control methods.

[0004] In order to achieve the above - mentioned objectives, the specific technical solutions of the present invention mainly include:

[0005] The present invention includes the following steps:

[0006] Step (1): Establish a nonlinear dynamic model of the deep - sea hydraulic manipulator;

[0007] Step (2): Design a differential tracker to obtain the speed estimation signal of the deep - sea hydraulic manipulator;

[0008] Step (3): Design an engineering mapping adaptive law to solve the problem of parameter uncertainty existing in the dynamic model of the deep - sea hydraulic manipulator;

[0009] Step (4) designs a finite-time convergent observer based on the non-linear dynamic model of the deep-sea hydraulic manipulator to observe the lumped disturbance of the system.

[0010] Step (5) designs a super-twisting sliding mode controller based on the backstepping control strategy, combined with the obtained speed estimation signal, engineering mapping adaptive law, and finite-time convergent observer, so as to obtain the valve port control law of the servo valve of the deep-sea hydraulic manipulator and achieve the precise motion control of the deep-sea hydraulic manipulator.

[0011] Compared with the existing technologies, the beneficial effects of the present invention are as follows:

[0012] 1. The method of the present invention establishes a super-twisting sliding mode control law for the deep-sea hydraulic manipulator including non-linear model parameter compensation, unknown disturbance compensation, and uncertainty parameter suppression, thereby overcoming the problems of high non-linearity and strong uncertainty accompanied in the process of establishing the controller, enhancing the robustness of the controller while improving the control accuracy of the deep-sea hydraulic manipulator, and improving the operation ability and control performance of the deep-sea hydraulic manipulator.

[0013] 2. The present invention obtains the unmeasurable speed signal of the deep-sea hydraulic manipulator by establishing a differential tracker; in addition, an engineering mapping adaptive law is constructed to overcome the problem of parameter uncertainty existing in the non-linear dynamic model of the deep-sea hydraulic manipulator; at the same time, a finite-time convergent observer is established, which can observe the disturbance amount at each joint of the manipulator during the movement of the deep-sea hydraulic manipulator and compensate it in the controller through feedback control, improving the robustness and control performance of the controller. Description of the Drawings

[0014] Figure 1 It is the system block diagram of the super-twisting sliding mode control method for the deep-sea hydraulic manipulator based on the finite-time convergent observer involved in the embodiment of the present invention.

[0015] Figure 2 It is the comparison diagram of the tracking errors generated during the control process of the super-twisting sliding mode control method and the PID control method for the deep-sea hydraulic manipulator based on the finite-time convergent observer involved in the embodiment of the present invention. Detailed Embodiments

[0016] In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the drawings and embodiments. The specific embodiments described in the present invention are only used to explain the present invention, but are not limited to the present invention, and the technical features involved can be combined with each other without conflict.

[0017] The present invention proposes a super-twisting sliding mode control method for a deep-sea hydraulic manipulator based on an observer, which mainly includes the following steps: First, for the mechanical configuration and hydraulic transmission principle of the deep-sea hydraulic manipulator, considering the mechanical characteristics and hydraulic dynamic characteristics, a nonlinear dynamic model of the deep-sea hydraulic manipulator is established; Second, for the case where there is no speed sensor on the deep-sea hydraulic manipulator, a differential tracker is designed to obtain the speed signal of the deep-sea hydraulic manipulator; Then, an engineering mapping adaptive law is designed to solve the problem of parameter uncertainty of the deep-sea hydraulic manipulator; In addition, based on the established nonlinear dynamic model of the deep-sea hydraulic manipulator, a finite-time convergent observer is designed to observe the lumped disturbance of the system in real time; Finally, based on the backstepping control strategy, the speed estimation value, the unmeasurable lumped disturbance estimation quantity and the adaptive law estimation value of the deep-sea hydraulic manipulator are fed back to the controller, and a super-twisting sliding mode controller is designed to form a complete closed-loop control system of the deep-sea hydraulic manipulator, realizing the precise position control of the deep-sea hydraulic manipulator.

[0018] Therefore, the super-twisting sliding mode control method for a deep-sea hydraulic manipulator based on a finite-time convergent observer proposed by the present invention can effectively reduce the negative impact of the nonlinearity and uncertainty of the deep-sea hydraulic manipulator during the movement process on the control accuracy in the case of no speed sensor, and solve the problem of insufficient control accuracy of the deep-sea hydraulic manipulator existing in the existing control methods.

[0019] Now in combination with Figure 1 make a further description of the present invention:

[0020] Step (1), considering the mechanical configuration of the deep-sea hydraulic manipulator, hydraulic dynamic characteristics, and multi-source interference factors such as parameter uncertainty and unknown disturbances from the outside world in the system, establish a nonlinear dynamic model of the deep-sea hydraulic manipulator.

[0021] The nonlinear dynamic model of the deep-sea hydraulic manipulator is mainly composed of the dynamic relationship of the link dynamics model of the deep-sea hydraulic manipulator, the nonlinear dynamics model of the hydraulic system, and the nonlinear model of the servo valve orifice.

[0022] (1.1) Establish the link dynamics model of the deep-sea hydraulic manipulator, specifically as follows:

[0023]

[0024] where q = [q1, q2,..., q n T , and respectively represent the position, speed and acceleration of the joints of the deep-sea hydraulic manipulator, and n represents the number of degrees of freedom of the deep-sea hydraulic manipulator. M ∈ R n×n represents the symmetric positive definite inertia matrix of the deep-sea hydraulic manipulator, C ∈ R​n×n represents the Coriolis and centrifugal force matrix of the deep - sea hydraulic manipulator, \(G\in R\) n×1 represents the gravity matrix of the deep - sea hydraulic manipulator, \(H\in R\) n×n represents the hydrodynamic matrix of the deep - sea hydraulic manipulator, including viscous resistance and added - mass effect. \(\tau\) represents the torque vector applied at each joint of the deep - sea hydraulic manipulator; \(D\) represents the system disturbances such as friction and interference suffered by the manipulator during movement.

[0025] (1.2) Establish a non - linear dynamic model of the hydraulic system as follows:

[0026] The joints of the deep - sea hydraulic manipulator involved in the present invention all achieve joint movement by controlling servo valves. Define \(y = diag[y_1,y_2,\cdots,y\) n represents the moving distances of the hydraulic cylinders at different joints, where \(diag[]\) represents a skew - symmetric matrix. Assume that there is no internal leakage in the hydraulic system during the movement of the deep - sea hydraulic manipulator, then the dynamic equation of the hydraulic system can be modeled as:

[0027]

[0028] where, \(V\) 01 and \(V\) 02 respectively represent the initial volumes of the oil - inlet chamber and the oil - return chamber of the hydraulic cylinders at each joint of the deep - sea hydraulic manipulator; \(\beta\) e represents the effective bulk modulus of the hydraulic oil; \(P_1=[P\) 11 ,P\) 12 ,\cdots,P\) 1n T and \(P_2=[P\) 21 ,P\) 22 ,\cdots,P\) 2n T respectively represent the pressures of the oil - inlet chamber and the oil - return chamber of the hydraulic cylinders at each joint of the deep - sea hydraulic manipulator; and respectively represent the pressure change rates of the oil - inlet chamber and the oil - return chamber; \(A = diag[A_1,A_2,\cdots,A\) n represents the effective stamping areas of the oil - inlet chamber and the oil - return chamber of the hydraulic cylinders at each joint of the deep - sea hydraulic manipulator; is the complete differential matrix between the moving distance y of each hydraulic cylinder and the position q of each joint, and can be written as:

[0029]

[0030] Meanwhile, the torque generated by the hydraulic system at each joint of the deep - sea hydraulic manipulator is:

[0031] ​​

[0032] In addition, Q1 and Q2 represent the fluid flow rates of the oil inlet chamber and the oil return chamber of the hydraulic cylinder at each joint of the deep-sea hydraulic manipulator, and their relationship with the servo valve input signal is as follows:

[0033]

[0034] where k q represents the proportional coefficient between the servo valve input signal and the flow rate; g1(P1, u) and g2(P2, u) respectively represent the non-linear relationships between the control signal of the servo valve and the fluid flow rates of the oil inlet chamber and the oil outlet chamber of the hydraulic cylinder. In the formula

[0035]

[0036] where, k q represents the proportional coefficient between the control input signal of the servo valve and the flow rate of the hydraulic cylinder chamber; C d represents the flow coefficient of the throttle orifice of the servo valve; w represents the area gradient of the throttle orifice; ρ represents the density of the hydraulic oil.

[0037] (1.3) Establish a non-linear model of the servo valve orifice as follows:

[0038] The relationships g1(P1, x v ) and g2(P2, x v ) between the spool displacement of the servo valve and the flow rates of the oil inlet chamber and the oil outlet chamber of the hydraulic cylinder are as follows:

[0039]

[0040] where P s and P r are the oil supply and oil return pressures of the hydraulic system in the deep-sea hydraulic manipulator respectively. The definition of the function L(Ξ) in formula (7) is as follows:

[0041]

[0042] This invention mainly considers a type of servo valve with fast response characteristics. The bandwidth of the servo valve is much higher than the actual bandwidth of the system. Therefore, the high-order dynamic response characteristics of the servo valve can be ignored, and the dynamic characteristics of the servo valve can be approximated as a proportional link, that is, the spool displacement x v of the servo valve is proportional to the control input u as follows:

[0043] x v = k v u (9)

[0044] where k v represents the proportional gain of the servo valve. Therefore, combining formula (7), formula (5) is rewritten as:

[0045]

[0046] where k Q = k q k v is the total flow gain.

[0047] (1.4) The non-linear dynamic model of the deep-sea hydraulic manipulator is as follows:

[0048] Based on the dynamic models of the various subsystems of the deep-sea hydraulic manipulator in steps (1.1)-(1.3), considering the non-linearity and uncertainty problems of the deep-sea hydraulic manipulator, the state variables of the deep-sea hydraulic manipulator are defined where x3 represents the thrust formed by the pressure difference between the inlet cylinder and the return cylinder of the deep-sea hydraulic manipulator, and the non-linear dynamic model of the deep-sea hydraulic manipulator is established as follows:

[0049]

[0050] where φ1(x1,x2) = M(x1) -1 (C(x1,x2)+H(x2)), φ2 = M(x1) -1 , φ3 = M(x1) -1 G(x1), θ3 = βek q , f1 = M -1 D + Π1 and f2 = Π2; Π i is the process error generated during the modeling process, i = 1, 2.

[0051] Step (2) Design a differential tracker to obtain the speed estimation signal of the deep-sea hydraulic manipulator; the specific process is as follows:

[0052] The differential tracker has the advantages of simple structure and not over-amplifying system noise. Therefore, during the controller design process, a differential tracker is used to obtain the speed estimation signal of the joints of the deep-sea hydraulic manipulator, and its specific structure is as follows:

[0053]

[0054] where the position signal x1 of the joints of the deep-sea hydraulic manipulator is used as the input signal of the differential tracker; m0 is the state estimator related to the position of the differential tracker, m1 is the state estimator related to the speed of the differential tracker; l1 is the gain coefficient related to the position error of the differential tracker, l2 is the gain coefficient related to the speed error of the differential tracker; v is the output signal of the differential tracker, that is, the speed estimation signal of the joints of the deep-sea hydraulic manipulator.

[0055] Step (3) Design the engineering mapping adaptive law to solve the problem of parameter uncertainty in the dynamic model of the deep-sea hydraulic manipulator;

[0056] Design the engineering mapping adaptive law, and estimate the uncertain parameters in the dynamic model of the deep-sea hydraulic manipulator through the online update of the engineering mapping adaptive law. The specific establishment process is as follows:

[0057] For the convenience of subsequent explanation, the following is defined here denotes the estimated value of parameter ·, and the estimation error is defined as All subsequent relevant parameter forms conform to this definition. Define Θ = [θ1, θ2, θ3] T and, in order to ensure that the estimated value is within a certain range, design the engineering mapping adaptive law as shown below:

[0058]

[0059] where Γ = diag{Γ1, Γ2, Γ3}, representing a positive definite adaptive gain matrix; the projection mapping is defined as follows:

[0060]

[0061] In addition, the adaptive law vector N ∈ R 3 is defined as follows:

[0062]

[0063] where ψ = diag[ψ1, ψ2, ψ3] is a positive definite matrix.

[0064] Step (4) Based on the non-linear dynamic model of the deep-sea hydraulic manipulator, design a finite-time convergent observer to observe the lumped disturbance of the system;

[0065] Based on the established non-linear dynamic model of the deep-sea hydraulic manipulator, establish a finite-time convergent observer for the deep-sea hydraulic manipulator. By using the velocity estimation signal of the deep-sea hydraulic manipulator joint as the observer input, the lumped disturbance estimation quantity at each joint of the deep-sea hydraulic manipulator is obtained. The specific process is as follows:

[0066] For generality, reconstruct the non-linear dynamic model of the deep-sea hydraulic manipulator to obtain the following form:

[0067]

[0068] where, denotes the polynomial matrix containing indefinite parameters in the non-linear dynamic model of the deep-sea hydraulic manipulator, x i(x, φ) represents the part without undetermined parameters in the non - linear dynamic model of the deep - sea hydraulic manipulator, and f i represents the system disturbance, where i = 1, 2. In order to estimate the disturbance in the non - linear dynamic model of the deep - sea hydraulic manipulator, a virtual system of the deep - sea hydraulic manipulator is constructed as follows:

[0069]

[0070] Among them, represents the state quantity of the virtual system of the deep - sea hydraulic manipulator, represents the input signal of the virtual system of the deep - sea hydraulic manipulator. According to formulas (16) and (17), the following estimation error expression is obtained:

[0071]

[0072] When , the control signal of the virtual system of the deep - sea hydraulic manipulator is the lumped disturbance estimator in the non - linear dynamic model of the deep - sea hydraulic manipulator. Therefore, the control signal of the virtual system of the deep - sea hydraulic manipulator is constructed as follows:

[0073]

[0074] where k f1 is the gain coefficient of the estimation error, k f2 is the gain coefficient of the estimation error non - linear term, ω1 is the gain coefficient of the estimation error in the disturbance estimator, ω2 is the gain coefficient of the disturbance adaptive term in the disturbance estimator; r is a positive constant satisfying r > 2; η i is the adaptive term. In addition, define According to formula (4 - 9), the expression of the change rate of the estimation error is as follows:

[0075]

[0076] Step (5) Based on the back - stepping control strategy, design a super - twisting sliding - mode controller to achieve the precise motion control of the deep - sea hydraulic manipulator.

[0077] According to the non - linear dynamic model of the deep - sea hydraulic manipulator, combined with the velocity estimation signal, engineering mapping adaptive law, and finite - time convergence observer obtained in steps 2 - 4, based on the back - stepping control strategy, design a super - twisting sliding - mode controller to obtain the virtual control thrust, and then obtain the valve - port control law of the servo valve of the deep - sea hydraulic manipulator to achieve the precise motion control of the deep - sea hydraulic manipulator. The specific process is as follows:

[0078] Define the trajectory tracking error z1 = x1 - x 1dand the virtual control law for the trajectory tracking error x 1d is the desired trajectory tracking error signal; k1 is a positive definite diagonal matrix; α1 is the virtual control law with respect to x1, which is convenient for subsequent controller design. Define the variable z2 as follows for subsequent analysis and calculation:

[0079]

[0080] Taking the derivative of z2 in formula (21) gives:

[0081]

[0082] Define the integral sliding mode surface of the variable z2 as follows:

[0083]

[0084] where c1 is the gain coefficient of the error z2, and s(0) represents the value of the sliding mode surface at the initial moment. Taking the derivative of the sliding mode surface (23) gives:

[0085]

[0086] Based on formula (24), the velocity estimation signal v obtained in step 2, the estimated value of the engineering mapping adaptive law obtained in step 3 and the lumped disturbance estimator obtained in step 4 Design the virtual control thrust α2 as follows:

[0087]

[0088] where α 2a represents the adaptive term in the virtual control thrust α2, α 2s represents the robust term in the virtual control thrust α2, α aux represents the auxiliary term in the virtual control thrust α2, specifically as follows:

[0089]

[0090] where k 11 is the gain coefficient of the sliding mode term in α 2s k 12 is the gain coefficient of the sliding mode nonlinear term in α 2s k 13 is the gain coefficient of the sign function term in α aux and is the filtering gain coefficient in α aux In addition, define the thrust error z3 = x3 - α2, and take the derivative of the thrust error z3 to get:

[0091]

[0092] However, since the speed signal x2 of the deep-sea hydraulic manipulator joint is an unmeasurable state, it is difficult to directly perform differential operations during the process of solving the differential of the virtual control thrust α2. To solve this practical problem and avoid amplifying system noise, the differential operation of the virtual control thrust α2 is defined to include the following two parts:

[0093]

[0094] where represents the differentiable part in the virtual control thrust α2, represents the non-differentiable part caused by disturbances and other uncertain factors, which will be suppressed during the subsequent controller design process.

[0095] Define the integral sliding mode surface of the thrust error z3 as follows:

[0096]

[0097] where c2 is the gain coefficient of the thrust error z3. Differentiating the sliding mode surface (29) gives:

[0098]

[0099] Based on formula (27), the velocity estimation signal v obtained in step 2, the estimated value of the engineering mapping adaptive law obtained in step 3 and the lumped disturbance estimator obtained in step 4 the following valve port control law u for the final deep-sea hydraulic manipulator servo valve is obtained:

[0100]

[0101] where u a represents the adaptive term in the control law u, u s represents the robust term in the control law u, u aux represents the auxiliary term in the control law u, specifically as follows:

[0102]

[0103] where k 21 is the gain coefficient of the sliding mode term in u a , k 22 is the gain coefficient of the sliding mode nonlinear term in u a , k 23 is the gain coefficient of the sign function term in u aux , is the filtering gain coefficient in u aux .

[0104] Based on the design of the above-mentioned super-twisting sliding mode controller, by using the speed estimation signal v of the deep-sea hydraulic manipulator, the estimated value of the engineering mapping adaptive law lumped disturbance estimator and the position information x1 of the joints of the deep-sea hydraulic manipulator obtained by the sensor and the thrust x3 formed by the pressure difference between the inlet cylinder and the return cylinder of the deep-sea hydraulic manipulator are fed back to the super-twisting sliding mode controller in real time. According to the obtained valve port control law of the servo valve of the deep-sea hydraulic manipulator, the control of the deep-sea hydraulic manipulator is realized, thereby forming a complete closed-loop control system for the deep-sea hydraulic manipulator, achieving the precise control of the deep-sea hydraulic manipulator, and solving the problem of insufficient control accuracy of the existing deep-sea hydraulic manipulator.

[0105] Finally, in order to prove the effectiveness of the above control method, an experimental verification of the super-twisting control method for the deep-sea hydraulic manipulator based on a finite-time convergence observer was carried out on a two-degree-of-freedom deep-sea hydraulic manipulator experimental platform, and through comparison and verification with the common PID control method, the effectiveness and practicability of the control method proposed by the present invention were verified.

[0106] During the control process, the control gain coefficients of PID are as follows: k p = diag[75, 225], k I = diag[25, 10]; the control gain coefficients of the control method proposed by the present invention are as follows: k1 = diag[50, 45], k 11 = diag[500, 300], k 12 = diag[200, 170], k 13 = diag[100, 100], k 21 = diag[2000, 1500], k 22 = diag[770, 650], k 13 = diag[500, 400], Γ = diag[5.0×10 -4 , 4.2×10 -4 , 0.08, 0, 05, 20, 35]; k f1 = diag[100, 80, 50, 55], k f2 = diag[50, 60, 35, 35], r = diag[3, 2, 2, 4], ω1 = diag[110, 100, 65, 70], ω2 = diag[225, 210, 175, 190].

[0107] The results of the trajectory tracking error generated by the experiment are as Figure 2As shown, where the C1 controller represents the control method proposed in the present invention, and C2 represents the common PID control method. Through Figure 2 It can be seen that the control method proposed in the present invention can exhibit good control performance in the motion control process of deep-sea hydraulic manipulators with strong uncertainty and high nonlinearity, and accurately control the joint angle to follow the desired signal. In addition, under the action of the C2 controller, the tracking error of the deep-sea hydraulic manipulator exceeds 0.05 rad, while the error of the C1 controller is less than 0.03 rad, which proves the excellence and effectiveness of the controller proposed in the present invention. At the same time, it can change well with the change of the speed of the desired signal during the motion process, has smaller control error and faster response time compared with the common PID control method, reflecting that the super-twisting sliding mode control method of the deep-sea hydraulic manipulator based on the finite-time convergent disturbance observer established in the present invention has better transient response performance and better robustness, and can achieve precise control of the deep-sea hydraulic manipulator without a speed sensor, improving the control performance of the deep-sea hydraulic manipulator.

[0108] The above content is only the technical idea of the present invention, and the protection scope of the present invention cannot be limited thereby. Any modification made on the basis of the technical solution according to the technical idea proposed in the present invention falls within the protection scope of the claims of the present invention.

Claims

1. A super-helical sliding mode control method for a deep-sea hydraulic manipulator based on an observer, characterized in that: The steps include: Step (1) establishing a nonlinear dynamic model of a deep-sea hydraulic manipulator; Step (2) designing a differential tracker to obtain a velocity estimation signal of the deep-sea hydraulic manipulator; Step (3) designing an engineering mapping adaptive law to solve the parameter uncertainty problem existing in the deep-sea hydraulic manipulator dynamics model; Step (4) Based on the nonlinear dynamic model of the deep-sea hydraulic manipulator, a finite-time convergence observer is designed to observe the lumped disturbance of the system; Step (5) is based on the inversion control strategy, combining the obtained velocity estimation signal, the engineering mapping adaptive law and the finite time convergence observer to design a super helical sliding mode controller, thereby obtaining the valve port control law of the deep-sea hydraulic manipulator servo valve and realizing precise motion control of the deep-sea hydraulic manipulator.

2. The super-helical sliding mode control method of a deep-sea hydraulic manipulator based on an observer according to claim 1 is characterized in that: The nonlinear dynamic model of the deep-sea hydraulic manipulator is composed of the dynamic relationship between the deep-sea hydraulic manipulator connecting rod dynamic model, the hydraulic system nonlinear dynamic model and the servo valve port nonlinear model. (1.1) Establish the dynamic model of the deep-sea hydraulic manipulator connecting rod, as follows: where q=[q1,q2,...,q n ] T , and Respectively represent the position, velocity and acceleration of the deep-sea hydraulic manipulator joint, n represents the number of degrees of freedom of the deep-sea hydraulic manipulator; M∈R n×n represents the symmetric positive definite inertia matrix of the deep-sea hydraulic manipulator, C∈R n×n represents the Coriolis and centrifugal force matrices of the deep-sea hydraulic manipulator, G∈R n×1 Represents the gravity matrix of the deep-sea hydraulic manipulator, H∈R n×n represents the fluid dynamics matrix of the deep-sea hydraulic manipulator, including viscous drag and added mass effect; τ represents the torque vector applied to each joint of the deep-sea hydraulic manipulator; D represents the system disturbance to which the manipulator is subjected during movement; (1.2) Establish the nonlinear dynamic model of the hydraulic system, as follows: The joints of the deep-sea hydraulic manipulator are all controlled by servo valves to achieve joint movement; define y = diag[y1, y2, ..., y n ] represents the moving distance of the hydraulic cylinder at different joints, where diag[] represents a skew-symmetric matrix; the dynamic equation of the hydraulic system is modeled as: Among them, V 01 and V 02 Respectively represent the initial volumes of the oil inlet chamber and the oil return chamber of the hydraulic cylinder at each joint of the deep-sea hydraulic manipulator; β e Represents the effective bulk modulus of hydraulic oil; P1=[P 11 ,P 12 ,...,P 1n ] T and P2=[P 21 ,P 22 ,...,P 2n ] T Respectively represent the pressure of the oil inlet chamber and the oil return chamber of the cylinder at each joint of the deep-sea hydraulic manipulator; and Respectively represent the pressure change rate of the oil inlet chamber and the oil return chamber; A=diag[A1,A2,...,A n ] represents the effective stamping area of ​​the oil inlet chamber and the oil return chamber of the oil cylinder at each joint of the deep-sea hydraulic manipulator; Distance traveled for each hydraulic cylinder y The position of each joint q The complete differential matrix between can be written as: At the same time, the torque generated by the hydraulic system at each joint of the deep-sea hydraulic manipulator is: In addition, Q1 and Q2 represent the fluid flow rate of the oil inlet chamber and oil return chamber of the hydraulic cylinder at each joint of the deep-sea hydraulic manipulator, and their relationship with the servo valve input signal is as follows: where k q represents the proportional coefficient between the servo valve input signal and the flow rate; g1(P1,u) and g2(P2,u) represent the nonlinear relationship between the servo valve control signal and the fluid flow rate of the oil inlet and outlet chambers of the oil cylinder, respectively; Among them, k q Indicates the proportionality coefficient between the control input signal of the servo valve and the flow rate of the hydraulic cylinder chamber; C d represents the flow coefficient of the throttle hole of the servo valve; w represents the area gradient of the throttle hole; ρ represents the density of the hydraulic oil; (1.3) Establish a nonlinear model of the servo valve port, as follows: The relationship between the servo valve core displacement and the flow rate of the oil inlet and outlet chambers of the oil cylinder is g1(P1,x v ) and g2(P2,x v )as follows: Where P s and P r are respectively the oil supply and return pressures of the hydraulic system in the deep-sea hydraulic manipulator; the function L(Ξ) in formula (7) is defined as follows: Ignoring the high-order dynamic response characteristics of the servo valve, the dynamic characteristics of the servo valve are approximated as a proportional link, that is, the valve core displacement x of the servo valve v It is proportional to the control input u, as follows: x v =k v u(9) Where k v The table is the proportional gain of the servo valve; therefore, combined with formula (7), formula (5) is rewritten as: Among them, k Q =k q k v is the total flow gain; (1.4) The nonlinear dynamic model of the deep-sea hydraulic manipulator is as follows: According to the dynamic model of each subsystem of the deep-sea hydraulic manipulator in steps (1.1)-(1.3), considering the nonlinearity and uncertainty of the deep-sea hydraulic manipulator, the state variables of the deep-sea hydraulic manipulator are defined. Where x3 represents the thrust formed by the pressure difference between the oil inlet cylinder and the oil return cylinder of the deep-sea hydraulic manipulator. The nonlinear dynamic model of the deep-sea hydraulic manipulator is established as follows: Where φ1(x1,x2)=M(x1) -1 (C(x1,x2)+H(x2)),φ2=M(x1) -1 ,φ3=M(x1) -1 G(x1), f1=M -1 D+Π1 and f2=Π2; Π i is the process error generated in the modeling process, i=1,2.

3. The super-helical sliding mode control method of a deep-sea hydraulic manipulator based on an observer according to claim 1 is characterized in that: The specific process of step (2) is as follows: In the controller design process, a differential tracker is used to obtain the velocity estimation signal of the deep-sea hydraulic manipulator joint. Its specific structure is as follows: Among them, the position signal x1 of the deep-sea hydraulic manipulator joint is used as the input signal of the differential tracker; m0 is the state estimate of the differential tracker related to the position, and m1 is the state estimate of the differential tracker related to the speed; l1 is the gain coefficient of the differential tracker related to the position error, and l2 is the gain coefficient of the differential tracker related to the speed error; v is the output signal of the differential tracker, that is, the speed estimation signal of the deep-sea hydraulic manipulator joint.

4. The super-helical sliding mode control method of a deep-sea hydraulic manipulator based on an observer according to claim 1 is characterized in that: The specific process of step (3) is as follows: To facilitate the subsequent explanation, we define denotes the estimated value of parameter · and defines the estimation error The subsequent related parameter forms all conform to this definition; define Θ=[θ1,θ2,θ3] T , and to ensure that the estimated value Within a certain range, the following engineering mapping adaptive law is designed: Where Γ=diag{Γ1,Γ2,Γ3}, represents the positive definite adaptive gain matrix; the projection mapping is defined as follows: In addition, the adaptive law vector N∈R 3 , defined as follows: Where ψ=diag[ψ1,ψ2,ψ3] is a positive definite matrix.

5. The observer-based super-helical sliding mode control method for deep-sea hydraulic manipulator according to claim 1, characterized in that: The specific method of step (4) is as follows: Based on the established nonlinear dynamic model of the deep-sea hydraulic manipulator, a finite-time convergence observer of the deep-sea hydraulic manipulator is established. The velocity estimation signal of the deep-sea hydraulic manipulator joint is used as the observer input to obtain the lumped disturbance estimation at each joint of the deep-sea hydraulic manipulator. The specific process is as follows: In order to maintain generality, the nonlinear dynamic model of the deep-sea hydraulic manipulator is reconstructed and obtained as follows: in, represents the polynomial matrix containing uncertain parameters in the nonlinear dynamic model of deep-sea hydraulic manipulator, χ i (x, φ) represents the part of the nonlinear dynamic model of the deep-sea hydraulic manipulator without uncertain parameters, f i Represents system disturbance, i = 1, 2; In order to estimate the disturbance in the nonlinear dynamic model of the deep-sea hydraulic manipulator, a deep-sea hydraulic manipulator virtual system is constructed as shown below: in, Represents the state quantity of the deep-sea hydraulic manipulator virtual system, represents the input signal of the deep-sea hydraulic manipulator virtual system; according to formulas (16) and (17), the following estimation error expression is obtained: when When the control signal of the deep-sea hydraulic manipulator virtual system is It is the lumped disturbance estimate in the nonlinear dynamic model of the deep-sea hydraulic manipulator; therefore, the control signal of the deep-sea hydraulic manipulator virtual system is constructed. as follows: where k f1 is the gain coefficient of the estimation error, k f2 is the gain coefficient of the estimation error nonlinear term, ω1 is the gain coefficient of the estimation error in the disturbance estimation term, ω2 is the gain coefficient of the disturbance adaptive term in the disturbance estimation term; r is a positive constant, satisfying r>2; η i is an adaptive term; in addition, the definition According to formula (4-9), the rate of change of the estimated error is expressed as follows:

6. The observer-based super-helical sliding mode control method for a deep-sea hydraulic manipulator according to any one of claims 1 to 5, characterized in that: The specific method of step (5) is as follows: According to the nonlinear dynamic model of the deep-sea hydraulic manipulator, combined with the speed estimation signal obtained in step 2-step 4, the engineering mapping adaptive law and the finite time convergence observer, based on the inversion control strategy, the super-helical sliding mode controller is designed to obtain the virtual control thrust, thereby obtaining the valve port control law of the deep-sea hydraulic manipulator servo valve, and realizing the precise motion control of the deep-sea hydraulic manipulator; the specific process is as follows: Define the trajectory tracking error z1 = x1-x 1d and the virtual control law for trajectory tracking error x 1d is the desired trajectory tracking error signal; k1 is a positive definite diagonal matrix; α1 is the virtual control law about x1; the variable z2 is defined as follows: By taking the derivative of z2 in formula (21), we can obtain: The integral sliding surface of variable z2 is defined as follows: Where c1 is the gain coefficient of error z2, s(0) represents the sliding surface value at the initial moment; the derivative of the sliding surface (23) is obtained: Based on formula (24), the velocity estimation signal v obtained in step 2, and the estimated value of the engineering mapping adaptive law obtained in step 3 And the lumped disturbance estimate obtained in step 4 The virtual control thrust α2 is designed as follows: where α 2a represents the adaptive term in the virtual control thrust α2, α 2s represents the robust term in the virtual control thrust α2, α aux represents the auxiliary term in the virtual control thrust α2, as follows: where k 11 is α 2s The gain coefficient of the sliding mode term, k 12 is α 2s The gain coefficient of the sliding mode nonlinear term, k 13 is α aux The gain coefficients and is α aux The filter gain coefficient in; In addition, the thrust error z3 is defined as x3-α2, and the thrust error z3 is derived to obtain: However, since the velocity signal x2 of the deep-sea hydraulic manipulator joint is unmeasurable, it is difficult to directly perform differential operations in the process of solving the differential of the virtual control thrust α2. In order to solve this practical problem and avoid amplifying the system noise, the differential operation of the virtual control thrust α2 is defined to include the following two parts: in represents the differentiable part of the virtual control thrust α2, represents the non-differentiable part due to disturbances and other uncertainties; The integral sliding surface of the thrust error z3 is defined as follows: Where c2 is the gain coefficient of the thrust error z3; the derivative of the sliding surface (29) is obtained: Based on formula (27), the velocity estimation signal v obtained in step 2 and the estimated value of the engineering mapping adaptive law obtained in step 3 are And the lumped disturbance estimate obtained in step 4 The final valve port control law u of the deep-sea hydraulic manipulator servo valve is as follows: where u a represents the adaptive term in the control law u, u s represents the robust term in the control law u, u aux represents the auxiliary term in the control law u, as follows: Among them, k 21 for u a The gain coefficient of the sliding mode term, k 22 for u a The gain coefficient of the sliding mode nonlinear term, k 23 for u aux The gain coefficient of the sign function term in , for u aux The filter gain coefficient in ; Based on the design of the super-helical sliding mode controller, the velocity estimation signal v of the deep-sea hydraulic manipulator and the estimated value of the engineering mapping adaptive law are transformed into Lumped disturbance estimator The position information x1 of the deep-sea hydraulic manipulator arm joint obtained by the sensor and the thrust x3 formed by the pressure difference between the oil inlet cylinder and the oil return cylinder of the deep-sea hydraulic manipulator arm are fed back to the super-helical sliding mode controller in real time. The deep-sea hydraulic manipulator arm is controlled according to the valve port control law of the deep-sea hydraulic manipulator arm servo valve, thereby forming a complete deep-sea hydraulic manipulator arm closed-loop control system and realizing precise control of the deep-sea hydraulic manipulator arm.

Images