An improved multiple model particle filtering approach for manoeuvring target tracking using Airborne GMTI with geographic information

摘要

This paper proposes a novel ground vehicle tracking method using an airborne ground moving targetudindicator radar where the surrounding geographic information is considered to determine vehicle’sudmovement type as well as constrain its positions. Multiple state models corresponding to differentudmovement modes are applied to represent the vehicle’s behaviour within different terrain conditions.udBased on geographic conditions and multiple state models, a constrained variable structure multipleudmodel particle filter algorithm aided by particle swarm optimisation is proposed. Compared withudthe traditional multiple model particle filtering schemes, the proposed algorithm utilises a particleudswarm optimisation technique for the particle filter which generates more effective particles andudgenerated particles are constrained into the feasible geographic region. Numerical simulation resultsudin a realistic environment show that the proposed method achieves better tracking performanceudcompared with current state-of-the-art ones for manoeuvring vehicle tracking.