A Kalman Filter with a Turning Acceleration Estimation Using Mode Probability

概要

論文の詳細を見る
For multi-target tracking, the a-fl filter and the Kalman filter, both of which don't require computational requirements, have been studied. The IMM(Interactive Multiple Model) algorithm has been proposed. The IMM is expected to reduce tracking errors for both non-maneuvering and maneuvering target. However, IMM requires heavy computational burden, because it utilizes multiple Kalman filters in parallel. On the other hand, the Kalman filter with a turning acceleration estimator, which realizes real-time opera- tion, has been proposed. However, the Kalman filter with a turning acceleration estimator has a problem of setting threshold in the maneuver detector. In this paper, we propose the hybrid filter with a linear filter and a turning acceleration estimation filter without a maneuver detector. The proposed filter integrates the estimations using the normalized likelihood of each filter. The proposed method can prevent deteriorating tracking accuracy by reducing the probability of miss maneuver detection when a target maneuvers. We evaluate the performance of the proposed filter by computer simulation, and show the effectiveness of the proposed filter, comparing with the conventional Kalman filter and the Two-Stage Kalman filter.
社団法人電子情報通信学会の論文
2002-03-01