Incorporating Multi-Source Transfer Learning into an Unscented Kalman Filter for Object Tracking

Omar Alotaibi, Brian L. Mark

Abstract: We develop a framework for incorporating Bayesian transfer learning into an unscented Kalman filter (UKF) to track a nonlinear dynamic motion model in a multi-source system consisting of two or more sensors. This framework extends a previous UKF-based transfer learning scheme for object tracking in a two-sensor system. Our objective is to extend the application of transfer learning to leverage knowledge gained from several source sensors. To achieve this aim, we generalize the transfer learning approach to online multi-source Bayesian transfer learning. The extended framework is approximated via a UKF, where the predicted observation densities of one or more sources are transferred to a primary sensor experiencing higher measurement noise intensity. By leveraging knowledge from multiple sources instead of a single source, the tracking performance of the primary sensor is able to achieve a higher level of estimation accuracy. The performance gain increases proportionally with the number of source sensors. Our numerical results demonstrate the effectiveness of multi-source Bayesian transfer learning in improving tracking accuracy.