| Popis: |
In this paper, we introduce a target Bearing and Heading Tracking algorithm (BHT) using bearing-only measurement from a stationary sonar sensor. Since the target location is impossible to be obtained from a single stationary bearing-only sensor, the optimal output of this sensor is the quality bearing tracks. The conventional Bearing-Only Tracker (BOT) using an Extended Kalman Filter (EKF) does not perform well. This is because EKF is sensitive to the initial state, and the initial state is difficult to estimate in bearing-only detection. In BHT algorithm, a target heading is computed from a sequence of measured bearings, and this heading form the measurement vector together with measured bearing. The state vector is the same as the extended Kalman Filter to track target dummy location and velocity, and the final bearing and heading values are computed from tracked location and velocity. The simulation results show that BHT algorithm is with good performance in terms of track accuracy, and with less false tracks. Besides the quality bearing tracks, the BHT also provides heading information which can be further utilized, as a very useful feature, on target localization in multiple sensors environment. |
