Distributed Multi-sensor Multi-target Tracking with Random Sets II

摘要

An approach for distributed multi-sensor multi-target tracking with random sets is introduced. For each sensor, probability hypotheses density filter is employed to obtain a state estimate set, then, weighted tracks association is used to correlate the state estimates. Experiments show this approach to be able to estimate both the number of tracked objects, as well as the states of the objects, robustly from noisy observations.