Multi-Ellipsoidal Extended Target Tracking With Variational Bayes Inference

In this work, we propose a novel extended target tracking algorithm, which is capable of representing or group targets with multiple ellipses. Each ellipse modeled by an unknown symmetric positive-definite random matrix. The proposed model requires solving two challenging problems. First, the data association problem between measurements and sub-objects. Second, inference that involves non-conjugate priors likelihoods needs to be solved within recursive filtering framework. We utilize variational Bayes method solve approximate intractable true posterior. performance solution demonstrated in simulations real-data experiments. results show our outperforms state-of-the-art methods terms accuracy lower computational complexity.