International Conference on Information Technology and Computer Science, 3rd (ITCS 2011)
119 Research on Maneuvering Target Tracking for Passive Sensors Based on Truncation Gauss Probability Model
Download citation file:
- Ris (Zotero)
- Reference Manager
Because the two-dimensional passive sensors can only receive the azimuth from target without location, so the observability is hardly satisfied when single passive sensor is used for maneuvering target tracking. In this paper, the dimension of measurement equation based on Gauss-Hermite filter for single passive sensor is extended, and the model based on Gauss-Hermite filter for multi-sensor is established. The Singer model assumes a uniform probability distribution on the target acceleration which is independent of x and y direction, and it is the same with constant velocity target and constant acceleration target. For the intense maneuvering target, it will induce serious error. The truncation Gauss probability model is a non-zero-mean correlative model, and it can more factually reflect maneuvering range and the change of intensity of target, so it is an effective model at present. Based on the truncation Gauss probability model, the algorithm for maneuvering target tracking with Gauss-Hermite filter for passive multi-sensor is presented. Simulation results show that this method can steadily track the maneuvering target tracking.