Estimating positions and paths of moving objects

Abstract

Several applications require support for representing and analyzing moving objects. Such applications include wildlife tracking, emergency dispatch, vehicle navigation, fleet management, storm tracking, and military applications to name a few. Enhanced data collection technologies such as high resolution satellite imagery, videogrammetry and GPS in combination with wireless communications are rapidly increasing the feasibility of obtaining information on moving objects and fueling new research and development. In these applications there is a need to efficiently answer questions about moving objects such as where they are at a specific time, where they have been in the past, where they will be in the future, and their relationships to static or other moving objects. Complete knowledge of object movements is not possible but movements can be predicted with some degree of reliability. Positions and paths of moving objects can be estimated from a set of observations. This paper reports on an approach that uses movement profiles, and movement histories in addition to observations to more reliably estimate positions and paths of moving objects.