Information-Theoretic Multi-Robot Adaptive Exploration and Mapping of Environmental Hotspot Fields
Kian Hsiang, John M. Dolan · 2009
Recent research in robot exploration and mapping has focused on sampling hotspot elds. This exploration task is formalized by [3] in a decision-theoretic planning framework called MAXP. The time complexity of solving MAXP approximately depends on the map resolution, which limits its use in large-scale, high-resolution exploration and mapping. To alleviate this computational diculty, this paper presents an informationtheoretic approach to MAXP (iMAXP); by reformulating the cost-minimizingiMAXP as a reward-maximizing problem, its time complexity becomes independent of map resolution and is less sensitive to increasing robot team size. Using the reward-maximizing dual, we derive a novel adaptive variant of maximum entropy sampling, thus improving the induced policy performance. We also demonstrate the superior performance of exploration policies for sampling the log-Gaussian process to that of policies for the Gaussian process in mapping the hotspot eld. Lastly, we provide sucient conditions