Stochastic Motion Planning Under Partial Observability for Mobile Robots With Continuous Range Measurements

In this article, we address the problem of stochastic motion planning under partial observability, more specifically, how to navigate a mobile robot equipped with continuous range sensors, such as LIDAR. contrast many existing robotic methods, explicitly consider uncertainty state by modeling system partially observable Markov decision process (POMDP). Recent work on general purpose POMDP solvers is typically limited discrete observation spaces, and does not readily apply proposed due measurements from build upon an Monte Carlo tree search method, (POMCP), propose new algorithm POMCP++. Our can handle spaces novel measurement selection strategy. The POMCP++ overcomes overoptimism in value estimation rollout policy removing implicit perfect assumption at phase. We validate theory proving it algorithm. Through comparisons other methods that also be applied problem, show yields significantly higher success rate total reward.