Adaptive Vo-Vo filter for maneuvering targets with time-varying dynamics

The paper addresses the problem of tracking multiple highly dynamic targets. The Vo-Vo filter (also known as Generalized Labeled Multi-Bernoulli or GLMB filter) is an attractive approach which possesses useful analytical properties and is a closed form solution to the Bayes multi-target filter. However, Vo-Vo filter with a fixed motion model is not general enough to accommodate maneuvering targets. In this paper, we propose an adaptive Vo-Vo filter for maneuvering targets with time-varying dynamics. The explicit prediction and update equations of the proposed adaptive Vo-Vo filter are derived and implemented using the Gaussian Mixture (GM) approximations. Simulation results in two challenging tracking scenarios prove the effectiveness of the proposed algorithm.