This work presents a method for motion planning under uncertainty to deal with situations where ambiguous data associations result in a multi-modal hypothesis on the robot state.
We present an approach to plan actions that sequentially disambiguate a multimodal belief to achieve a unimodal belief in finite amount of time. Experimental results are provided using a simulation of a non-holonomic ground robot operating in an artificial maze-like environment.
We demonstrate two experiments wherein the robot is given no a priori information about its initial pose and the planner is tasked with localizing the robot.