Probabilistic self-localization for sensor networks

Abstract

This paper describes a technique for the probabilistic self-localization of a sensor network based on noisy inter-sensor range data. Our method is based on a num-ber of parallel instances of Markov Chain Monte Carlo (MCMC). By combining estimates drawn from these parallel chains, we build up a representation of the un-derlying probability distribution function (PDF) for the network pose. Our approach includes sensor data incre-mentally in order to avoid local minima and is shown to produce meaningful results efficiently. We return a distribution over sensor locations rather than a single maximum likelihood estimate. This can then be used for subsequent exploration and validation.