Maneuvering Target Tracking in a Cluttered Environment with Variable Demension Filter

A new technique for tracking a maneuvering target in a cluttered environment is developed. This approach does not rely on a statistical description of the maneuver as a random process. Instead, the state model for the target is changed when a maneuver is detected. Undesired measurements due to clutter or false alarms are assumed to occur uniformly and independently distributed. The Probabilistic Data Association Filter (PDAP) is used for both state models of the target. Measurement sequences over a window are tested for occurrence of possible maneuver. In the process, tracks are formed, deleted and merged. Simulation has shown the effectiveness of the scheme.