A new measurement partition for extended target tracking based on CFSFDP algorithm
Luo-jia Chi, Xinxi Feng, Miao Lu · 2017
In an extended targets tracking, especially in clutter environment with unknown and varying number of targets, measurement partition is of great importance for accurate filtering. But it is always computational and complicated in many cases. To solve this problem, a new measurement set partitioning based on CFSFDP algorithm is proposed. Firstly, the Local Outlier Factor is used to reduce the clutter, then a new method based on density peaks inis used to partition the measurement set. This algorithm is not sensitive to the shape and can realize accuracy measurement set partitioning for any shape of targets. Simulation results show that in the case of target crossover, the proposed algorithm can ensure the performance of the extended target tracking, as well as reduce the computational time effectively..