Side scan sonar grid map for Unmanned Underwater Vehicle navigation

摘要

Unmanned Underwater Vehicles (UUVs) are popular tools nowadays in marine exploration and oceanographic research. How to navigate an UUV in unknown or unstructured environments without long baseline or short baseline information are always a challenging problem. Side Scan Sonar (SSS) are common payloads on UUVs for navigation and environment detection. This study proposes a real time SSS seafloor mapping method combining the motion information and the observation model of the sonar to localize the UUVs. The working principle of SSS, sonar footprint region calculation and the acoustic velocity in water are considered during the feature localization procedure. An occupancy grid mapping algorithm is then used to map the seafloor combining the uncertainties of UUV's motion and measurement errors from the real time observations. Scanlines' information instead of the SSS images during the landmarks' position detection process was implemented on an UUV and experimental results conducted in a sandy seabed with concrete targets are demonstrated.