Transformer Based Hybrid Bernoulli GSF Tracking Algorithm

Target tracking in the maritime domain is a challenging problem due to several factors such as fluctuating and spiky sea clutter that is non-Gaussian with low target signal-to-interference ratios (SIR). The authors propose a method which uses machine learning to improve the representation of targets in a cluttered background. Tracking is then performed on point measurements obtained from the machine learning algorithm. The tracking algorithm selected is the Bernoulli Gaussian Sum filter. The error performance of the proposed method is presented and compared to an existing tracking algorithm applied to the maritime domain.