Merging Probabilistic Observations for Mobile Distributed Sensing

Abstract

We present a method for representing, communicating, and fusing multiple noisy observations of an object made by multiple distributed robots. The approach relies on re-parameterization of the canonical two-dimensional Gaussian distribution to a form that corresponds more naturally to the observation space of a robot. The approach enables two or more observers to achieve greater effective sensor coverage of the environment and improved accuracy in object position estimation. We demonstrate empirically that, when using our approach, more observers achieve more accurate estimations of an object's position. Quantitative evaluations of the technique in use on mobile robots are provided. The method is applied in three application areas, including object location, object tracking, and ball position estimation for robotic soccer. Preliminary qualitative results from these sample applications are discussed.