Reactive Combination of Belief Over Time Using Direct Perception

One issue for autonomous mobile robots operat ing in unknown, or part ial ly known, domains is how to handle uncertainty in their sensor observations over t ime. Methods such as probablistic belief networks and survivor functions are generally unsatisfactory because they require explicit models of the robot's interactions w i th its environment, including possible contravening events. This information is difficult to obta in, and is philosophically incompatible w i th reactive behaviors.