Multi-bernoulli sensor control using Cauchy-Schwarz divergence
Amirali Khodadadian Gostar, Reza Hoseinnezhad, Alireza Bab‐Hadiashar · Figshare · 2024
This paper presents an information theoretic sensor-control method to be used within a multi-Bernoulli filter for multi-target tracking. The proposed sensor-control method follows an information-theoretic approach, in which a divergence measure between predicted and updated multi-target densities is maximized. The divergence function of choice in this paper is the Cauchy-Schwarz divergence. Via first moment approximation, a closed form solution for Cauchy-Schwarz divergence in a multi-Bernoulli filtering scheme is derived. Simulation results demonstrate that the proposed method significantly outperforms recently developed information-theoretic sensor-control methods in terms of computational cost while returning comparable tracking accuracy.