Auv Slam And Experiment Using A Mechanical Scanning Forward-looking Sonar

Navigation technology is one of the most important challenges in theapplications of AUV which navigates itself in the complex undersea environment.The ability of localizing a robot and accurately mapping its surroundingssimultaneously, namely the simultaneous localization and mapping (SLAM) problem,is a key prerequisite of truly autonomous robots.In this paper a modified-FastSLAM algorithm is proposed and used in thenavigation of C-Ranger, which is an open-frame AUV developed as our researchplatform. A mechanical scanning imaging sonar is chosen as the active sensing for theunderwater vehicle. The modified-FastSLAM implements the update relying on theon-board sensors of C-Ranger. On the other hand, the algorithm employs the dataassociation which combines the single particle maximum likelihood method withnegative evidence method, and uses the rank-based resampling to overcome theparticle depletion problem. Moreover, a compensation method based on feedback ofthe AUV pose is presented to overcome distortion of acoustic image due to thevehicle motion. In order to verify the feasibility of the proposed methods forC-Ranger, simulation experiments and sea trial in Tuan Dao Bay are conducted. Theexperimental results show the modified-FastSLAM employed for the navigation ofC-Ranger AUV is actually more effective and accurate while compared with thetraditional method.