286 results about "State space equation" patented technology

The invention relates to an automatic guided vehicle (AGV) adaptive path tracking method based on MPC. The method comprises the following steps: S1, establishing a kinematic discretization error model based on the kinematic characteristics of the two-wheel differential AGV, processing a continuous kinematic equation by adopting an Euler discretization method, expressing a dynamic change relationship between a transverse deviation distance and an angle deviation in a state-space equation form, and generating a kinematic discrete state-space model; the method has the advantages that the continuous equation is processed by establishing the kinematics discretization error model and adopting the Euler discretization method, the deviation dynamic relation is expressed through the state space, the calculation process is simplified, the precision is kept, sensor data are fused, the wheel type odometer, IMU and laser data are integrated through the extended Kalman filtering algorithm, and the precision is improved. The real-time position and angle deviation are calculated, the positioning accuracy is improved, a model prediction controller objective function is designed, a weight matrix and boundary constraint are introduced according to constraint conditions, and the effect of obstacle avoidance constraint is combined.

The present disclosure discloses a method for attitude control of a quadrotor UAV, comprising establishing an attitude dynamics model of the quadrotor UAV, establishing a motion equation and a state-space equation of a UAV control system, determining an LADRC-CFO, and establishing a differential tracker for reducing a system overshoot.

The invention provides a delay compensation control method for a double-vibrator wave energy conversion device, which comprises the following steps of: performing numerical modeling and hydrodynamic analysis on the double-vibrator wave energy conversion device, establishing a time domain motion equation of the double-vibrator wave energy conversion device, and introducing a time delay function into the time domain motion equation to simulate control signal transmission delay; simulating the execution delay of the brake by using a partial differential equation; establishing a state-space equation of the device, replacing a convolution term of the time-domain motion equation with the state-space equation, and calculating to obtain a motion state of the device; a Hamiltonian function is defined to convert a constrained optimization problem into an unconstrained optimization problem, and the Hamiltonian function is solved to obtain an optimal control criterion considering control delay so as to realize maximization of energy capture under the optimal control criterion. According to the method, the operation characteristics of a physical system are truly reflected by introducing control delay, so that the locking control method is effectively implemented in an actual physical device, and the energy capture efficiency of a wave energy conversion device and the reliability of system operation are effectively improved.

The invention discloses a distributed driving electric vehicle path tracking and stability control method and system, and belongs to the technical field of electric vehicle active safety control. The method comprises the following steps: constructing a state-space equation based on a driver path tracking model and a vehicle three-degree-of-freedom motion model; establishing a dynamic envelope region based on a phase plane method, inputting the current parameters of the vehicle into the dynamic stable envelope region, and outputting a weight; constructing a controller based on the state-space equation, and taking the weight output by the dynamic stable envelope region as the control weight of the controller; inputting the vehicle state variables and the vehicle parameters into a controller, and outputting a longitudinal driving torque, an additional yaw torque, an additional front wheel turning angle and a driver intention turning angle required by driving; the additional front wheel turning angle and the driver intention turning angle are superposed to serve as input of a steering motor; a torque distributor is designed to distribute a longitudinal driving torque and an additional yaw torque, and the distributed torques are used as given values of the four hub motors.

The invention discloses a thermal power generating unit coordinated optimization control method based on model prediction, and particularly relates to the field of thermal power generating unit coordinated optimization control, and the method comprises the steps: S1, collecting the real-time operation parameters of a thermal power generating unit through a sensor network, carrying out the preprocessing of the real-time operation parameters, and building a unit operation state database; s2, constructing a machine-furnace dynamic prediction model based on the operation state database, adopting a state-space equation to describe and predict the parameter trend in the next 60s, and verifying the precision; s3, an optimization objective function containing multi-parameter deviation and energy consumption items is constructed according to the prediction trend, the optimal control quantity is solved in a rolling optimization mode every 10 s, and the control quantity is constrained by a rated parameter range; according to the method, the dynamic prediction model of the machine furnace is constructed and combined with the rolling optimization algorithm, the parameter change trend is predicted in advance to realize multi-variable cooperative control, and the advantages of reducing the parameter fluctuation range and improving the unit operation stability under the variable load working condition are achieved.

The invention provides a lithium battery parameter online identification method based on dynamic weight particle swarm optimization, and relates to the field of power lithium battery model parameter identification. According to the method, a lithium battery equivalent circuit model based on a second-order RC network is established, and a state-space equation is derived according to the Kirchhoff's current law and the voltage law; a discrete output model is obtained through Laplace transformation and discretization processing; constructing a recursive least square parameter updater with a forgetting factor, and jointly optimizing an initial parameter vector of the forgetting factor recursive least square and the forgetting factor by adopting a dynamic weight particle swarm optimization algorithm; and based on the optimized initial parameter vector and the forgetting factor, performing recursive least square updating with the forgetting factor, performing online identification on model parameters, and reversely deducing an equivalent parameter value in an original circuit. According to the method, the parameter identification precision can be remarkably improved, and the problem that data saturation parameter convergence is slow in a traditional recursive least square method is solved.

The invention discloses an energy storage battery SOC and SOH joint estimation method based on adaptive double extended Kalman filtering, and the method comprises the steps: constructing a second-order RC equivalent circuit model of an energy storage battery, and building a state-space equation of a battery state and a battery parameter; measuring the open-circuit voltage of the energy storage battery under different SOCs through a pulse discharge experiment, and fitting a polynomial function relationship between the SOC and the open-circuit voltage by using a least square method; establishing a battery capacity model and an internal resistance model for energy storage battery SOH estimation; the measurement noise variance is dynamically adjusted based on an improved Sage-Husa adaptive algorithm, and joint estimation of the SOC and the SOH of the energy storage battery is realized through an iteration mode. According to the method, a relatively large state estimation error caused by fixed measurement noise is avoided, self-adaptive adjustment of the measurement noise is realized, and the reliability and the accuracy of SOC and SOH joint estimation of the energy storage battery are improved.

The invention discloses an active disturbance rejection controller control method and device, and the method comprises the steps: receiving an actual output feedback signal of a controlled object and a historical output control signal of a controller based on a linear expansion state observer, calculating the total disturbance of a system in real time through a three-order state-space equation, and obtaining a disturbance estimation value; receiving the disturbance estimation value and the control input gain coefficient based on a compensator, calculating a dynamic compensation amount for offsetting the total disturbance of the system, and outputting the dynamic compensation amount to a signal synthesis node; a deviation signal of a set value signal used for defining the target state of a controlled object and a feedback signal actually output by the controlled object is received based on a controller, and an anti-interference basic control signal is generated in combination with the bandwidth of the controller; and subtracting the dynamic compensation amount from the anti-interference basic control signal based on the signal synthesis node to generate an active-disturbance-rejection control amount, outputting the active-disturbance-rejection control amount to a controlled object, and suppressing external interference and internal parameter drift of the controlled object.

The invention discloses a periodic disturbance compensation method and device for an input time delay system, equipment and a medium, and relates to the field of disturbance control, and the method comprises the steps: obtaining a state-space equation of the input time delay system; according to the state-space equation of the input time-delay system, converting the input time-delay system into a non-input time-delay system by adopting a time-delay compensation method; based on the error signal of the no-input time delay system, estimating the value of the to-be-determined parameter by adopting an internal model principle and a linear regression method to obtain an estimated value of the to-be-determined parameter; and determining a control input signal of the zero-input time-delay system by adopting the feedback controller according to an output signal of the zero-input time-delay system and the to-be-determined parameter estimation value so as to perform periodic disturbance compensation on the zero-input time-delay system. The feedback controller is adopted to perform feedback control on the control input signal according to the output signal of the system, and compensation of unknown time-varying periodic disturbance of the input time delay system can be realized.

The invention belongs to the technical field of robot cooperative control, and particularly relates to a distributed self-adaptive control method and system for cooperation of multiple mechanical arms, and the method comprises the steps: decoupling a pure internal force vector from the data of a six-dimensional force sensor at the tail end through pre-identified kinetic parameters; calculating a cooperative confrontation index according to the pure internal force vector and the relative position vector, calculating a self-adaptive internal force penalty weight through nonlinear mapping based on the cooperative confrontation index, establishing a state-space equation of the system based on a double-integral model, introducing an equivalent environment stiffness matrix, mapping a prediction error vector in a prediction time domain into a prediction internal force vector, and obtaining a prediction internal force vector; and a cost function containing the product of the self-adaptive internal force penalty weight and the predicted internal force vector is constructed, an optimal control increment sequence is obtained by minimizing the cost function, and a speed instruction is updated to drive the mechanical arm to move. According to the method, the robustness and the execution stability of the system are effectively improved, and the cooperation safety is guaranteed.

The invention relates to an automobile thermal management system control method based on multi-target model predictive control, and belongs to the technical field of automobile thermal management systems. The method comprises the steps of establishing a discrete state space model of the thermal management system, setting a state variable, a control variable, a disturbance variable and a state space equation of the discrete state space model of the thermal management system, and then establishing a multi-target cost function and constraint conditions with stability, safety and low energy consumption of the thermal management system as targets; and then, based on fuzzy control logic, combining with an actual working condition, dynamically calculating a weight of each cost in the multi-target cost function, and finally solving an optimization problem according to the multi-target cost function under the determined weight in a discrete state space model of the thermal management system to obtain an optimal control input. And the optimal control input is transmitted to an actuator to regulate and control the automobile thermal management system. The system can adapt to various working conditions and environments, battery safety and comfort of a passenger compartment can be guaranteed, and energy efficiency and the service life of the actuator can be both considered.

A trajectory planning method for a lower limb exoskeleton includes: determining a state space equation corresponding to the lower limb exoskeleton using dynamic analysis; determining an objective function for trajectory planning of the lower limb exoskeleton according to the state space equation, wherein a phase variable corresponding to the objective function is a time-related phase variable; and performing trajectory planning of the lower limb exoskeleton according to the objective function to obtain a first gait trajectory corresponding to the lower limb exoskeleton, wherein the first gait trajectory is a gait trajectory corresponding to the time-related phase variable.

The present disclosure provides a cooperative distributed model predictive control (Co-DMPC)-based chassis multi-agent system (MAS) cooperative control method for autonomous vehicles, a controller, and a storage medium. A distributed state-space equation with state coupling and control input coupling characteristics is established. Meanings and transformation methods of predicted trajectories, assumed trajectories, and optimal trajectories of the states and control inputs are designed, providing a communication basis for information exchange between the agents. In order to coordinate the global performance indexes of a vehicle, a local agent optimization problem considering cost coupling is established, and the influence of the cooperative relationship on the control effect is quantitatively analyzed through adaptive weight coefficients. A method of performing a plurality of iterations within a unit sampling time is adopted, and iteration errors are utilized to enable the controller to achieve a balance between solution accuracy and efficiency.

The invention discloses a suspension pitching cooperative control system and method combining dynamic vertical force distribution and multi-mode braking, and particularly relates to the field of vehicle suspensions. Meanwhile, the braking torque distribution unit and the suspension system are controlled in real time by defining a state machine; vehicle low-speed target braking torque and tire vertical force are calculated through the expected deceleration and correspondingly input into a braking torque distribution unit and a four-degree-of-freedom suspension model; converting the four-degree-of-freedom suspension model into a state-space equation and discretizing the state-space equation to obtain a prediction model, and further deriving a state prediction equation of the future N steps; designing a cost function comprising a state quantity weight matrix Q, an input quantity weight matrix R and a state final value weight matrix F, substituting a prediction state into the cost function to obtain an optimal input sequence, dynamically adjusting suspension rigidity, damping and a braking force distribution strategy, and correcting a control instruction in real time by using closed-loop feedback of a vehicle body attitude sensor, so as to improve the stability of a vehicle body. And the suspension pitching control is realized.

The invention discloses a grid-connected inverter control method based on an MAGRNN neural network observer, and aims to solve the problems that a model prediction control method is insufficient in state estimation precision and worsened in control performance under the conditions of model mismatch, parameter offset and power grid voltage distortion. For an LCL type grid-connected inverter, firstly, a prediction model is established based on a state-space equation; secondly, real-time estimation parameters of an inductance parameter state observer are constructed; the method comprises the following steps: firstly, estimating an interference compensation item through a neuron self-adaptive adjustment mechanism, then introducing an MAGRNN neural network as a dynamic error compensator, using input and output of the observer as MAGRNN input, estimating the interference compensation item through the neuron self-adaptive adjustment mechanism, and finally, feeding back the interference item predicted by the MAGRNN to a model prediction controller for interference compensation. By utilizing the strong nonlinear fitting capability of the MAGRNN, the state estimation precision under complex working conditions such as parameter change and power grid disturbance is remarkably improved, and the control performance of the system is effectively improved.

The invention discloses a nonlinear adaptive control system design method and device based on a pseudo-inverse matrix, and relates to the field of adaptive control, and the method comprises the steps: S1, constructing a controlled object state-space equation comprising matching uncertainty and non-matching uncertainty; non-matching uncertainty is converted into matching uncertainty through pseudo-inverse transformation; s2, introducing the state vector into a regression vector to obtain an expansion regression vector, and obtaining an uncertainty function based on the expansion regression vector; s3, constructing a state observer, and obtaining a state observation value of the controlled object; s4, taking the difference between the state observation value and the state actual value as the excitation of the adaptive law, and updating the parameters of the uncertain function; s5, constructing a self-adaptive control law by taking the estimated value of the uncertain function as input; and S6, filtering an output signal of the adaptive control law through L1 to obtain a final control law. Through the expansion regression vector comprising the state vector, accurate compensation of the uncertain function is realized, and the self-adaptive efficiency is improved.

The invention provides a PI control method and device for laser soldering based on observation compensation, a controller and a program product, and the method comprises the steps: building a laser soldering thermodynamic model containing a thermal disturbance term based on a lumped parameter method, and constructing a state-space equation; designing a state observer based on the state-space equation, and estimating a state variable of the state-space equation and external disturbance acting on the state variable in real time; calculating a feedforward control quantity according to a target temperature curve and the thermodynamic model, calculating a compensation control quantity according to a disturbance value estimated by an observer, and calculating a PI control quantity according to the deviation between the output temperature and the target temperature to obtain a total control quantity; and in each sampling period, the total control quantity is dynamically adjusted according to the deviation between the current welding spot temperature and the target temperature, so that the temperature error tends to zero. According to the method, the disturbance observation technology and the feedforward-PI control are combined, and high-precision and high-robustness control over the welding temperature is achieved.

The invention discloses an IGBT life prediction method based on adaptive unscented particle filtering of time sequence form perception optimization, and belongs to the technical field of health prediction and reliability engineering. The method aims at the problems that a traditional filtering algorithm is prone to falling into local optimum in a high-dimensional parameter space and prediction trajectory forms are discontinuous, and comprises the steps that firstly, an IGBT nonlinear degradation model is established through a state-space equation; performing Sigma point sampling and nonlinear propagation at a particle level by adopting an unscented transformation and particle filtering cooperation mechanism, and fusing process noise to realize dynamic state prediction; constructing a Gaussian likelihood function by using the deviation between a predicted value and an actual observed value to generate a system residual sequence; an FCD fusion index is innovatively introduced, the index fuses a Pearson correlation coefficient and a Frechet distance to synchronously quantify trend consistency and form similarity, and self-adaptive genetic optimization is driven accordingly; and repeating the behaviors before a preset termination condition is met, and continuously updating the parameters of the state-space equation.

The invention discloses a multi-unmanned aerial vehicle formation control method based on distributed model predictive control. The method comprises the following steps: describing a motion law of each unmanned aerial vehicle by adopting a state-space equation; modeling a static obstacle as a geometric body, and setting a minimum safe distance; considering position tracking cost, control energy cost, input change cost and formation cost, and establishing an objective function for each unmanned aerial vehicle; adding a CBF security constraint according to whether the security domain has an obstacle avoidance demand or not; constructing an optimal control problem for each unmanned aerial vehicle; in each control period, the optimal control problem is solved; only the first item in the control sequence is adopted as an actual control instruction; and then the system state is updated based on the state-space equation, and prediction and optimization are carried out again in the next period. According to the invention, the problems of obstacle avoidance and formation keeping when the multi-unmanned aerial vehicle formation passes through the obstacle in a complex environment are effectively solved. The method can be widely applied to the field of unmanned aerial vehicle control.

The invention relates to the technical field of electromagnetic transient simulation, in particular to an electromagnetic transient simulation decoupling method, system and device based on a single-impedance branch and a medium, and the method comprises the steps: dividing a system to be simulated into two subsystems which are connected through the single-impedance branch, and carrying out the equivalent reconstruction to form a reconstructed branch model; establishing a discrete state space equation based on the reconstructed branch model; according to the main step length, the number of sub-steps meeting the stability condition is determined based on a small-step synthesis algorithm, and the main step length is equally divided into differential step lengths; in each main step length interval, performing iterative calculation on the state-space equation by using a differential step length to obtain a branch current and a capacitor voltage at the next moment; and injecting the calculated branch current serving as a known current source into an original network equation, solving in parallel to obtain the port voltage of the subsystem, and completing decoupling calculation and promoting simulation. According to the method, effective decoupling of the universal single-impedance branch can be realized, and the applicability, the numerical stability and the calculation efficiency of parallel simulation are improved.

The invention discloses a lithium ion battery SOC (state of charge) and SOH (state of health) joint estimation method based on an FOMISUKF-EKF algorithm. The method comprises the following steps: acquiring dynamic response and static characteristic data required by battery modeling; selecting a fractional order equivalent circuit model and establishing a state-space equation; identifying each parameter value of the model by adopting an ant colony algorithm; after the identified initial parameters are injected into the algorithm, SOC estimation under a micro scale is realized based on fractional order unscented Kalman filtering and a multi-innovation sliding mode observer; and when a preset period is met, switching to a macroscale, taking the SOC sequence as input, and performing joint identification on the battery parameters and SOH by adopting an EKF algorithm. According to the method, the multi-information theory and the sliding-mode observation algorithm are introduced on the basis of the traditional UKF algorithm, the sliding-mode observer is used as the state variable of the input end estimation system, the influence of system parameter change and external disturbance can be reduced to a certain extent, and rapid convergence and effective suppression of measurement noise are realized.

The invention relates to the technical field of power system stability analysis, in particular to an evaluation method and system for a stable weak node of a new energy grid-connected system.The method comprises the steps that node voltage serves as an input variable, node injection current serves as an output variable, and a linearization state-space equation of the new energy grid-connected system is constructed; obtaining a state matrix, an input matrix and an output matrix; calculating characteristic values of the new energy grid-connected system by using the state matrix, selecting a mode corresponding to a complex characteristic value with a real part absolute value smaller than a threshold value as a key mode, and calculating left and right characteristic vectors corresponding to the mode; using the input matrix, the output matrix, the left feature vector and the right feature vector to calculate a residue vector of the system corresponding to each node at the key mode; and calculating an L2 norm of the residue vector corresponding to each node, wherein the node corresponding to the residue vector of which the L2 norm is greater than a set threshold is the stable weak node of the system. According to the method, the most critical stable weak node in the system can be clearly identified.

The invention discloses a tracked vehicle loading wheel dynamic load analysis method, system and device and a medium, and relates to the field of tracked vehicle dynamics, and the method comprises the steps: constructing a road surface unevenness time domain model considering front and rear wheel input variable time difference correlation and left and right wheel input coherence, and outputting a road surface excitation input signal; establishing a parameterized kinetic model of the tracked vehicle suspension system and further determining a state-space equation; according to the road surface excitation input signal and the state-space equation, obtaining a dynamic load and a dynamic load coefficient of each loading wheel; establishing a tracked vehicle action system simulation model, performing dynamic load simulation based on different driving speeds, different road surface grades and different track shoe parameters by adopting a multi-body dynamics simulation method, and determining a change rule of a dynamic load of a loading wheel; the structure of the tracked vehicle loading wheel is optimized according to the change rule of the loading wheel dynamic load, the loading wheel dynamic load can be accurately analyzed, and a reliable basis is provided for design optimization of the tracked vehicle.

The invention discloses a structure parameter-load input joint identification method based on a Gaussian potential force model, and the method comprises the steps: obtaining a mass matrix, a stiffness matrix and a damping matrix, and constructing a state-space equation; modeling unknown load input into a zero-mean stationary Gaussian process with a specific covariance function to fuse prior information of the covariance function, and converting into an equivalent state-space equation; obtaining sparse dynamic response characteristics, and establishing a state space measurement equation; a structure state and load input are augmented into a state equation, the state equation and a measurement equation set are synthesized into a complete Gaussian process potential force model, unknown structure parameters are regarded as hyper-parameters, kernel function hyper-parameter optimization and structure parameter identification are achieved with the criterion of a maximized negative likelihood function, sequence reasoning is conducted on the Gaussian process potential force model, and the Gaussian process potential force model is obtained. Estimation of load input is obtained. The method has the advantages of being stable in parameter recognition result, high in precision, stable and reliable in load estimation, higher in robustness to model errors and measurement noise and the like.

This invention discloses a design method for a submarine vertical plane motion controller based on a closed-loop gain shaping algorithm. Based on this algorithm, the submarine depth controller and pitch controller designed according to the state-space equations of the submarine dynamics model can quickly and accurately achieve variable depth and constant depth maneuvers, optimizing the response time and steady-state error of the submarine's vertical plane motion controller, thus improving the submarine's underwater performance and safety. By constructing the depth and pitch nonlinear functions of the submarine's vertical plane motion controller, and combining these functions with the control laws of the bow and stern rudder actuators obtained from the vertical plane motion controller, the linear error feedback of the submarine's vertical plane motion controller is improved, enhancing the controller's robustness and energy-saving effect, reducing the impact of external interference, and decreasing the energy consumption of the servo motors.

The invention discloses a train rail coupling vibration control method, device and equipment and a readable storage medium, and is applied to the technical field of intelligent control, and the method comprises the steps: obtaining a nonlinear model considering a vibration mode of a rail based on a high-speed maglev train structure of a lap joint structure, a rail vibration equation and an electromagnet kinetic equation; determining motion characteristics of the nonlinear model to obtain a motion state space equation; determining a reference model corresponding to the nonlinear model, and determining a control law when the difference between the motion state space equation and the reference model is minimum; and determining target vibration control parameters by using a control law by taking the nonlinear model as a control object reference model as an ideal target. According to the method, the nonlinear model considering the vibration mode of the track is obtained through the high-speed maglev train structure based on the lap joint structure, the corresponding reference model and control law are designed, and due to the fact that the vibration mode of the lap joint structure is considered in the nonlinear model, the suppression effect on coupling vibration of the high-speed maglev train track is better.

This invention discloses a video reconstruction method driven by neuromorphic signals based on state-space equations, comprising: 1. constructing a video reconstruction network based on state-space equations; 2. introducing a random window displacement Mamba module designed for the spatial characteristics of neuromorphic signals; 3. introducing a Hilbert-filled Mamba module designed for the spatiotemporal characteristics of neuromorphic signals; 4. training the hybrid super-resolution network through backpropagation and continuously optimizing it until the loss function converges. The resulting video reconstruction model is used to reconstruct the neuromorphic signals to be processed, thereby generating corresponding high-quality videos. This invention achieves efficient operation and excellent visual effects through the linear global modeling capability of state-space equations and targeted network module design.

The application relates to the field of circuit state space equation solving technology in the real-time simulation field, and discloses a state space equation discretization method for capacitor and inductor element characteristics, which comprises the following steps: A) establishing a state space differential equation of a circuit; B) establishing a characteristic equation of a capacitor and an inductor element, determining a numerical integration method used by each capacitor and inductor element in the circuit, and obtaining a discrete time domain characteristic equation by using the numerical integration method; C) combining the discrete time domain characteristic equation of step B) and the state space differential equation of step A) to obtain a state space discrete equation of the circuit; and D) cyclic simulation. The state space equation discretization method for capacitor and inductor element characteristics improves the speed and precision of real-time simulation of a circuit system by independently discretizing the characteristic equations of each capacitor and inductor element and optimizing the discretization process of the state space equation.

The invention belongs to the technical field of automobiles, and particularly relates to a ground vehicle low-calculation-power path tracking control method considering a coupling gradient. The method comprises the following steps: firstly, establishing a vehicle dynamics model and a kinematics model considering the coupling gradient through coordinate transformation and dynamics analysis, and ensuring the path tracking precision of an unmanned ground vehicle on the coupling gradient; secondly, an unmanned ground vehicle path tracking system considering actuator time lag is constructed, and state space equation discretization is carried out, so that the system can be used for model prediction control; and finally, a path tracking control method is constructed, Laguerre function fitting is utilized to replace a link of solving an optimal control sequence through quadratic programming in traditional model prediction control, and the computing power demand of the control method is remarkably reduced. Through simulation verification, the provided control method can effectively reduce the computing power of the control method on the premise of ensuring the path tracking accuracy of the unmanned ground vehicle.

The invention discloses a ship deep reinforcement learning safety control method for resisting a false data injection attack. The method comprises the following steps: establishing a second-order state-space equation considering the false data injection attack; designing a self-adaptive updating law of the sensor attack compensator according to the reconstruction error variable; an estimated value of the auxiliary adaptive network model is obtained through a constructed auxiliary neural network weight adaptive law; according to the reconstruction error variable, the estimated value of the auxiliary adaptive network model and the adaptive updating law of the sensor attack compensator, constructing a feedforward elastic controller; and according to the estimation of the optimal feedback controller, a course feedback adaptive law and a feedforward elastic controller, constructing a course controller for deep reinforcement learning of the ship. According to the method, the problem that cooperative security control of high-precision tracking and low-power-consumption optimal control cannot be taken into consideration when an existing USV control technology faces model uncertainty in a complex environment and a sensor network is subjected to security threats of false data injection attacks is solved.