Autonomous information driven search for a diffusive source in an unknown structured environment

The paper presents a framework for autonomous search for a diffusive emitting source of a tracer (e.g. aerosol, gas) in an unknown environment of randomly placed and shaped obstacles. Typically, the concentration measurements from the emitting tracer are sporadic, noisy and without directional information. The search domain is modelled by a finite two-dimensional square grid, whose links represent the traversable paths (for emitted particles and for the searcher). The searcher must simultaneously estimate the source parameters, the map of the search domain and its own location within the map. The solution is formulated in the sequential Bayesian framework and implemented as a Rao-Blackwellised particle filter with information driven motion control. Numerical results demonstrate the concept and its performance.