Dynamics Analysis And Motion Control Research Of Four-vector RO

Vector propulsion technology can change the thrust direction generated by the thruster,improve the flexibility of the thruster,and be applied to remotely operated vehicles(ROVs)to better control the posture and motion of the ROV.In this paper,based on the coupling drive of four vector thrusters,a four-vector propulsion based ROV is designed and constructed.Due to the structural characteristics of vector thrusters,this paper needs to re-establish the kinematics and dynamics model of the ROV,and design a new control system for the vector ROV based on the dynamics model.Then,simulation and real machine experiments are conducted on the vector ROV.The research content of this article includes:(1)Design a new four vector ROV based on vector propulsion and build a 3D model using Solidworks software.Then,kinematics and dynamics modeling was carried out.In the dynamic model,the derivation of the structure matrix of the ROV vector propeller is considered,as well as the uncertainties and ocean current interference factors of the model.The dynamic model of the vector ROV is obtained through reasonable simplification.(2)Aiming at the motion control problem of vector ROV under complex sea conditions,a dynamic control system based on vector propulsion is proposed.The control system is divided into a motion control module and a power distribution control module.This paper designs two control strategies based on the simplified dynamic model of vector ROV.One is nonsingular fast terminal sliding mode control based on a novel nonlinear disturbance observer(NFTSMCNDO),The other is double loop neural network PID sliding mode control based on a nonlinear high order observer(DLNNSMC).In the motion control module,the motion control module adopts a control method that combines these two control strategies with PID control laws,and uses Lyapunov functions to verify the stability and convergence of these two control strategies.The power distribution control module designs an optimization objective function and its solution based on the pseudo inverse method.The established 3D model of vector ROV is imported into the underwater animation simulation environment of Gazebo to verify the effectiveness and robustness of the proposed control system.(3)In order to prove the superiority of the proposed NFTSMC NDO controller in performance,it is compared with three other advanced non singular fast terminal sliding mode control algorithms by simulation.Experimental results show that the proposed NFTSMC-NDO controller has better anti ocean current interference performance and faster position tracking control response,and is more robust than the other three advanced control algorithms.At the same time,the DLNNSMC proposed in this paper is compared with two other advanced PID neural network sliding mode control algorithms on the new AUV model,and the DLNNSMC is applied to the fixed-point tracking control of vector ROV.The experimental results show that the proposed NFTSMC-NDO controller has better performance in trajectory tracking control,and in the simulation of fixed-point tracking control,the tracking effect of DLNNSMC is better than that of NFTSMC-NDO.(4)In this paper,a hardware control system is established for real machine experiments.The system consists of a bottom drive control board and an upper robot operating system(ROS),while the onshore ground station is a laptop computer.The bottom layer uses the STM32 development board for drive control,while the upper layer uses the Raspberry pie ROS platform for control.The onshore ground station can communicate and control the Raspberry pie.Through real machine verification testing,it can be found that the prototype can meet the design requirements,and has achieved good control results in the experiment.Finally,the work of this article is summarized and the future research directions are prospected.