Issues in multi-agent search: False positives and Bayesian map updates
Claire L. Walton, Sean P. Kragelund, Isaac I. Kaminer · OCEANS 2017 - Aberdeen · 2017
This paper explores the incorporation of false positives and Bayesian updates into the framework of optimal search. Optimal search is an open-loop optimal control framework developed by the US Navy to optimize search and rescue path planning for rapid sweep sensors such as sonar. Although useful as a pre-planning tool, this framework lacks features necessary for online planning, such as quantification of the role of detection events and subsequent probability map updates. This paper discusses these limitations and provides tools for implementing these features.