Sensor Fusion With Feedback for Robust Target Tracking in Autonomous Ground Vehicles Under Missing Measurements

In this article, sensor fusion with feedback is investigated for robust target tracking in autonomous ground vehicles subject to missing measurements. Missing measurements are inevitable in dynamic scenarios, which degrades accuracy of target tracking. A compensation strategy is designed to address missing measurements for robust target tracking. An asynchronous sensor fusion with feedback scheme is proposed to obtain accurate target information by fully using asynchronous measurements. An asynchronous local estimator is designed using feedback fusion estimates and compensation measurements at each asynchronous sampling time. A distributed fusion estimator is designed to achieve target fusion estimates using sequential fast covariance intersection fusion. Simulation and experiment results are given to show effectiveness of the sensor fusion with feedback for robust target tracking in autonomous ground vehicles.