Track-to-track association using reference topology in the presence of sensor bias

Abstract

Track-to-track association is a fundamental problem in multi-sensor data fusion, which is complicated by random errors, false alarms, missed detections, and most profoundly, individual sensor bias. The state-of-the-art approach is to deal with bias estimates and track-to-track association jointly. However, the complexity of this approach is infeasible in the presence of a large number of targets. Moreover, track-to-track association problems due to sensor bias are usually studied assuming translation bias only and ignoring azimuth bias and range bias. However, azimuth bias arises naturally in sensor measurements and has a larger influence than translation bias. In this paper, a novel approach called the "reference topology" method is developed to account for sensor bias. Relative coordinates instead of absolute coordinates are used, which are less sensitive to sensor bias. The computational complexity of the reference topology approach grows linearly with the target number so that it can be implemented in the presence of a large number of targets. Simulations show that traditional approaches using absolute coordinates deteriorate dramatically in the presence of sensor bias. In contrast, the reference topology approach achieves nearly optimal performance even as the sensor bias grows.