Value-Based Observation with Robot Teams (VBORT) Using Probabilistic Techniques

Abstract

We present a behavior-based approach for directing the movements of robot teams engaged in mapping target objects in their environment. The resulting paths of the robots optimize the vantage points for all the robots on the team, maximizing information gain. At each step, each robot selects a movement to maximize the utility (in this case, reduction in uncertainty) of its next observation. Trajectories are not guaranteed to be optimal, but team behavior serves to maximize the team's knowledge since each robot considers the observational contributions of teammates. The VBORT approach is evaluated in simulation by measuring the resulting uncertainty about target locations compared to that obtained by robots acting without regard to teammate locations and to that of global optimization over all robots for each single step. The qualitative behavior of the team is sensible and close to the single-step optimal set of trajectories.