Supervised Learning Based Online Filters for Targets Tracking Using Radar Measurements

In the field of the radar target tracking, the state filtering plays an important role in estimating the target state. One of the widely-adopted filter is the Bayesian filter, which requires the prior information and an accurate modeling of the real tracking scene. Thus, the matching degree of the dynamic model has a key impact on the state estimation accuracy of the Bayesian filter. However, the target dynamic and radar measurement models cannot be approximated perfectly in an unknown and complicated environment, and the state estimation accuracy of the model-based filter is limited. To address the limitations of the Bayesian filter, a supervised learning based online filter for target tracking is proposed in this paper. In the proposed filter, a mapping among the radar measurements is first established in the context of the polar coordinate system. Then, based on data-driven, the state filtering is directly implemented to obtain the state estimate by using this mapping relationship. As such, the prior information is not required in the proposed filter, hence the proposed filter inherits a good estimation accuracy in unknown and complicated environments. Finally, simulation experiments clarify the effectiveness of our proposed filter via comparing with the traditional filter.