Provably-Convergent Bayesian Source Seeking with Mobile Agents in Multimodal Fields

Abstract

We consider source-seeking tasks, where the goal is to locate a source using a mobile agent that gathers potentially noisy measurements from the emitted signal. Such tasks are prevalent, for example, when searching radioactive or chemical sources using mobile sensors that track wind-carried particles. In this work, we propose an iterative Bayesian algorithm for source seeking, especially well-suited for challenging environments characterized by multimodal signal intensity and noisy observations. At each step, this algorithm computes a Bayesian posterior distribution characterizing the source's location using prior physical knowledge of the observation process and the accumulated data. Subsequently, it decides where the agent should move and observe next by following a search strategy that implicitly considers paths to the source's most likely location under the posterior. We show that the trajectory of an agent executing the proposed algorithm converges to the source's location asymptotically with probability one. We validate the algorithm's convergence through simulated experiments of an agent seeking a chemical plume in a turbulent environment.