Pose-based slam with probabilistic scan matching algorithm using a mechanical scanned imaging sonar

摘要

This paper proposes a pose-based algorithm to solve the full SLAM problem\udfor an Autonomous Underwater Vehicle (AUV), navigating in an unknown and\udpossibly unstructured environment. The technique incorporate probabilistic scan\udmatching with range scans gathered from a Mechanical Scanned Imaging Sonar\ud(MSIS) and the robot dead-reckoning displacements estimated from a Doppler Velocity\udLog (DVL) and a Motion Reference Unit (MRU). The raw data from the sensors\udare processed and fused in-line. No priory structural information or initial pose are\udconsidered. The algorithm has been tested on an AUV guided along a 600m path\udwithin a marina environment, showing the viability of the proposed approach.