A multiple target measurement retrieval algorithm based on K-N eighborhood membership degree P-PHD filtering

Abstract

In the extraction of multiple target state by P-PHD filtering, the traditional K-Means clustering method may cause problems like extended clustering time and incorrect clustering for clusters with different sizes. To solve this problem, a new measurement extraction method based on K neighboring membership degree is proposed. In this method, the category of measurement of the target is estimated by likelihood relations between the measurement and the particle. The particle is then distributed to every actual measurement category of each estimation by K neighboring membership degree. On this basis, new particle set is formulated and target state can be extracted directly from the set. The simulation results reveal that the proposed method is with more stable retrieval precision and less time complexity.