Probabilistic data association for dynamic world modeling: a multiple hypothesis approach

Proposes a multiple hypothesis approach for building and maintaining a world model for an autonomous robot vehicle. Dynamic world modeling requires the integration of multiple sensor observations obtained from multiple vehicle locations at different times. A crucial problem in this interpretation task is the presence of uncertainty in the origins of measurements (data association uncertainty) as well as in the values of measurements (noise uncertainty). The extended Kalman filter (EKF) has seen widespread use in robotics for dealing with the latter problem. The multiple hypothesis filter combines the basic machinery of the EKF with a rigorous Bayesian probabilistic data association framework in which to address temporal and spatial data association uncertainty. For dynamic world modeling, the approach results in multiple world models at a given time step, each one representing a possible interpretation of all past and current measurements and each having an associated probability. A single unified world model can be constructed by integrating all of the hypotheses to form a single hypothesis.<>