Cellular Ontology as a Basis for Spatiotemporal Reasoning

Consider the problem of guiding a vehicle that has to move within severe spatial constraints. It may be a car that is getting into or out of a busy parking lot, where it has to cooperate with the movements of other cars, possibly even in the style of a 15-puzzle. It may also be a helicopter that moves near buildings or other obstacles, and that is required to maintain line of sight to some objects (stationary or mobile ones) on the ground. Although such problems may be viewed as purely spatial ones, they may also have a temporal aspect if it is required to minimize travel time, or to be at certain points at certain clock times. For most vehicles the spatial and temporal considerations must be interrelated, in particular since a small turning radius in a maneuver may severely restrict the velocity or the acceleration/deceleration during that maneuver. It is natural to appeal to spatiotemporal reasoning techniques for mastering such tasks, particularly when the environment where the vehicle moves is well structured, as is typically the case e.g. in built-up areas and along roads. However, our problem has some characteristics that are not usually met by existing methods for spatiotemporal reasoning in the literature: