USV’s Modeling And Heading Control Based On Data Mining

The motion system modeling and heading control are the core technologies of autonomous USV. Efficient control systems are very important to improve the observation effect of reconnaissance equipment and the precision of the weaponry system.Due to the inherent characteristics of nonlinearity and uncertainty in USV motion system, traditional mathematical modeling and heading control methods can not achieve the satisfactory performance of system requirement.In the USV modeling, the key is to accurately characterize the ship dynamic property parameters based on the inherent framework of the hydrodynamic model and the response model. For the limitation of the inherent framework, it often constructs a mismatched model.In the USV control, the motion model of USV has strong randomicity and nonlinearity under wind, waves, flows and other disturbances, which make it difficult to achieve the fast and precise control effect using traditional control methods.To slove the above problems, USV motion system modeling is carried based on data mining technology and fuzzy inference system, and the identification result is applied to the design of USV heading control feedforward compensation controller, further more the Lyapunov stability analysis is used to prove the boundedness of the tracking error of the closed-loop control system.Fuzzy modeling method based on data mining technology is a data driven method, which uses the sampling data of rudder angle and heading to generate USV’s motion system model directly, without considering the internal movement mechanism and the hydrodynamic factors of USV.In the compensation control method based on fuzzy nonlinear identification, the fuzzy rules are adapted in time to improve the nonlinear control ability by using Lyapunov stability theory. Compared with the traditional PID heading control method, simulation results validate the rapidity and effectiveness of the feedforward compensation control method.At last, MATLAB is the platform to achieve the above content, and GUI tools is used to provide a visual integration platform for each function. The specific features included:Calculation of unmanned ship parameter;Fuzzy identification for USV motion by using sampling data;PID heading control of USV motion model; the compensation control of USV heading control based on fuzzy nonlinear identification.