Long-Range UWB Positioning-Based Automatic Docking Trajectory Design for Unmanned Surface Vehicle

Currently, unmanned surface vehicle (USV) is a hot research topic in marine technology and has been gradually applied into practice in different fields. Accurate positioning measurement and autonomous navigation are two fundamental issues for practical applications. Ultra-wideband (UWB)-based positioning mechanism has been widely used in indoor scenes. However, there are few cases that UWB positioning method is applied to outdoor occasions at present, especially in the sea surface. Due to occasional unavailability of global positioning system (GPS) signal, this article proposes a novel long-range UWB-positioning-based automatic docking path planning algorithm for USV. First, a novel automatic docking framework based on long-distance UWB positioning method is investigated. Furthermore, a hybrid filtering method using median filter (MF) and extended Kalman filter (EKF) is introduced to smooth ranging value, so as to calculate precise positioning information and heading angle of USV in the UWB-based coordinate system. The relative position measuring device based on long-distance UWB positioning technology is well-designed to be applied in the USV. Finally, segmented Dubins curve is applied to plan automatic docking trajectory iteratively so as to satisfy USV's dynamic constraints. Experimental results verify that the proposed long-range UWB-based positioning method has higher accuracy than traditional GPS, and therefore, it is particularly suitable for USV positioning without GPS signals. In the Sea trial, with the aid of derived high-precision positioning measurement instrumentation, the controlled USV can be navigated back to the dockyard along designed iterative Dubins curve.