The Design Of Deep Sea Work-class ROV Control System Based On Zynq

The Remotely Operated Vehicle(ROV)working in the deep sea is an important tool for the development of marine resources.With the deepening of oil and gas exploitation in China,the ROV which can work in the thousands of meters deep sea will play an important role.In the paper,the ROV control system design,modeling methods,trajectory tracking robust controller design,local path planning are researched basing the ROV produced by my lab.Firstly,the control system of ROV is studied using the platform of Xilinx's Zynq.In this platform,the hardware system is built firstly,and several serial modules are designed by Verilog HDL and the simulation experiment is carried out through Modelsim.Secondly,the software system is built,the Linux system is started,the driver program is written and the application program is written in the Linux system.Secondly,the ROV modeling method is studied.According to the state variables of ROV in multiple coordinate systems,the kinematics model of six degrees of freedom of ROV space is obtained.According to the force analysis of ROV at sea bottom,the dynamic equation of six degrees of freedom in ROV space is obtained by Newton dynamics equation and momentum moment equation.Based on the structural characteristics and motion characteristics of ROV,the six degree of freedom model is simplified reasonably,which provides a simple mathematical model for the subsequent control algorithm design.Thirdly,a robust controller for ROV trajectory tracking is designed.According to the characteristics of 3D trajectory tracking of ROV,ignoring the influence of transverse inclination and longitudinal inclination,the ROV space six degree of freedom model is simplified into a spatial four degree of freedom model.Considering the state variables of four degrees of freedom as a whole,a command Filtered adaptive Backstepping controller based on nonlinear disturbance observer is designed.Through the system simulation and contrast experiment,it can be seen that the controller can realize the precise control of ROV 3D trajectory tracking,and has better control quality compared than the PID controller.Finally,the ROV local path planning method is studied.Through the analysis of the advantages and disadvantages of fuzzy control and neural network,the fuzzy neural network controller is constructed by combining the two complement.According to the requirements of ROV local path planning,a multi-layer network structure based on Takagi-Sugeno model is designed.By calculating the output value of each layer network,the learning algorithm of parameter self-tuning is deduced.Through the sample training,the algorithm canautomatically adjust the parameters and make the network converge quickly.The trained fuzzy neural network controller can be used in the local path planning of ROV.