Neural fuzzy inference network approach to maneuvering target tracking

摘要

In target tracking study, the fast target maneuver detecting and highly accurate tracking are very important. And it is difficult to be solved. For the radar/infrared image fused tracking system, a extend Kalman filter combines with a neural fuzzy irr[erence network to be used in maneuvering target tracking. The features related to the target maneuver are extracted from radar, infrared measurements and outputs of tracking filter, and are sent into the neural fuzzy inference network as inputs firstly, and then the target's maneuver inputs are estimated, so that, the accurate tracking is achieved. The simulation results indicate that the new method is valuable for maneuvering target tracking.