Collaborative multi-vehicle localization and mapping in marine environments

This paper explains an application scenario of collaborative multi-vehicle simultaneous localization and mapping algorithm (CSLAM) in a marine environment using autonomous surface crafts (ASCs) in order to validate its performance. The motivation behind this is that a team of ASCs can explore a marine environment more efficiently and reliably than a single ASC. However use of multiple ASCs poses additional scaling problems such as inter-vehicle map fusion, and data association which needs to be addressed in order to be viable for various types of missions. In this paper we first demonstrate the steps of extending the single vehicle extended kalman filter based simultaneous localization and mapping (EKF-SLAM) approach to the multi-vehicle case. Performance of the algorithm is first evaluated using simulations and then using real data extracted from actual sea trials conducted in the littoral waters of Singapore (Selat Puah) using two ASCs. GPS data is used to assess the accuracy of localization and feature estimations of CSLAM algorithm. The improvements that can be achieved by using multiple autonomous vehicles in oceanic environments are also discussed.