An Improved Adaptive ET-PHD Algorithm for Newborn Target Intensity

In the case of unknown new target intensity, the traditional extended target probability hypothesis density (ET-PHD) filtering algorithm has a poor tracking effect. In this paper, an ET-PHD filtering algorithm based on measurement driven adaptive new target intensity is proposed. The new target intensity function is adaptively generated by the measured values obtained each time. The survival target and the new target are propagating in the prediction and updating stages and implemented under the Gauss mixture framework. The simulation results show that the proposed algorithm has a great advantage over the target number estimation and the OSPA distance compared with the traditional ET-GM-PHD filtering algorithm and improves the tracking performance of the extended target PHD filter.