No abstract

No abstract

There have been a lot of studies addressing target-tracking problems, in which targets like aircraft and missiles can move freely in the air without hard spatial constraints. Tracking ground targets is a completely different case. Variable terrain structures not only limit the target's moving capability, but also degrade the quality of measurement data. This paper describes an exploratory research project which studied the tracking of a single ground target via traditional and atypical approaches. Traditional Kalman techniques taking into account the additional information provided by the ground restrictions in the tracking process, a road network in our study, were implemented. Additionally, another tracker using the Hidden Markov Model (HMM) with transition array was also developed under the same scenario. The results showed that Kalman techniques with available road information significantly outperform the conventional Kalman approaches in terms of longitudinal and transversal errors at the time when the target maneuvers. The proposed adaptive HMM tracker, composed of some regional HMM trackers, is not sensitive to transversal maneuvers, but may yield large longitudinal errors at the time when the target approaches the boundary of each subscenario.