An improved unscented Kalman filter and the application to heading control of micro unmanned surface vehicles

Abstract

The micro unmanned surface vehicles (MTJSV) can be disturbed by ocean environment easily when moving and is usually equipped with low precision sensors. Thus, the data collected by sensors has the disadvantage of high noise, low precision and inadequate information. So it is hard to be used in motion control system directly. To solve above problems, an improved unscented Kalman filter (ITJKF) algorithm is proposed and applied to heading control system of MTJSV. The method optimizes unscented Kalman filter (UKF) and enhances the capability of anti-disturbance under high-noise condition. This paper is based on Charlie MTJSV and carries out the simulation and contrast experiment on ITJKF, UKF and extended Kalman filter (EKF). The results indicate that compared with UKF and EKF, IUKF has the advantage of strong ability of anti-disturbance, high precision and good convergence property. The algorithm eliminates the adverse effect caused by environmental noise and model vibration effectively and also estimates real-time yaw rate, yaw acceleration and create the condition for heading control. Therefore, the accuracy and stability of the control system are improved considerably.