Information fusion and tracking of maneuvering targets with artificial neural network
Zhongliang Jing, Hong Li Xu, Zhou Xueqin · 2002
A novel algorithm for tracking manoeuvring targets is presented in this paper. This algorithm is implemented with a pair of parallel adaptive filters by the information fusion technique together with the current statistical model (CSM) and backpropagation (BP) neural network. In order to adapt to different cases of movement, BP network fuses all state information of both filters and adjusts the system variance for one of the filters according to the trained sample set. Computer simulation results show that this algorithm can successfully tracks manoeuvring targets over a wide range of conditions, and has a higher tracking precision.>