Occupancy Grid-Based AUV SLAM Method with Forward-Looking Sonar

Simultaneous localization and mapping (SLAM) is an active method for Autonomous Underwater Vehicle (AUV), it can mainly be used in unknown complex areas such as coastal water, harbors, wharfs. This paper presents a practical occupancy grid-based based on forward-looking sonar AUV. The algorithm uses extended Kalman filter (EKF) to estimate the AUV motion states. First, SLAM fuses data coming from navigation sensors predict Subsequently, novel particle swarm optimization genetic (PSO-GA) scan matching employed grid map, pose would correct prediction Lastly, estimated states update map. experimental results field have validated that proposed adaptable underwater conditions, accurate enough use ocean engineering applications.