A Generalized Labeled Multi-Bernoulli Filter Implementation using Gibbs Sampling
This work addresses efficiency improvements for multi-target tracking in signal processing, representing an incremental advancement in filter implementation.
The paper tackled the computational complexity of the generalized labeled multi-Bernoulli filter by proposing an efficient implementation that combines prediction and update into a single step and introduces a Gibbs sampling-based truncation technique, resulting in drastically reduced complexity as demonstrated in numerical studies.
This paper proposes an efficient implementation of the generalized labeled multi-Bernoulli (GLMB) filter by combining the prediction and update into a single step. In contrast to the original approach which involves separate truncations in the prediction and update steps, the proposed implementation requires only one single truncation for each iteration, which can be performed using a standard ranked optimal assignment algorithm. Furthermore, we propose a new truncation technique based on Markov Chain Monte Carlo methods such as Gibbs sampling, which drastically reduces the complexity of the filter. The superior performance of the proposed approach is demonstrated through extensive numerical studies.