A Cooperative Navigation Method of Multiple AUVs for Wide Seafloor Survey –First Performance Evaluation in Sea Environments–

Abstract

This paper reports the sea experimental results and the performance evaluation of a cooperative navigation method of multiple autonomous underwater vehicles (AUVs). In the method, AUVs take two roles: moving and landmark. Moving AUVs estimate their states (horizontal positions and heading angle) from their ground velocity, horizontal angular velocity, and acoustical positioning relative to a landmark AUV remaining stationary on the seafloor. A stochastic approach called a particle filter is adopted for state estimation. Each AUV can navigate with small drift over a wide area by alternating the landmark role. The method was implemented in two hovering type AUVs (AUV Tri-Dog 1 and AUV Tri-TON 1). The method was evaluated in sea environments in November 2013. The AUVs succeeded in a cooperative navigation, alternately becoming the landmark. It was verified that the implemented method had the ability to perform the wide seafloor survey by only AUVs through sea experiments and post-processing simulation using the experimental results.