Spatially Grounded Multi-Hypothesis Tracking of People

People tracking is an important yet challenging task for mobile robots operating in populated environments and interacting with humans. What makes this problem difficult is that human behavior is complex and hard to predict. However, motion of people, the rate at which people appear and where they appear are not random but strongly place-dependent and follow patterns that are engendered by the environment. In this paper we make use of such information for the purpose of people tracking. Concretely, we learn a probabilistic representation, called spatial affordance map, to spatially ground activity events acquired by observing people in the environment. This representation is a non-homogeneous spatial Poisson process for which we derive expressions for life-long Bayesian learning. We show how the spatial affordance map can be used to compute refined probability distributions over hypotheses in a multi-hypothesis tracker and to make better, place-dependent predictions of human motion. In experiments with real data from a laser range finder, we demonstrate how both extensions lead to more accurate tracking behavior. The system runs in real-time on a typical desktop computer.