Joint Sensor Registration and Tracking with Heterogeneous Sensors

The number of nodes in sensor networks is continually increasing, and maintain accurate track estimates inside their common surveillance region is a mission-critical necessity. Modern sensor platforms are likely to carry a range of Commercial Off The Shelf (COTS) sensors with different modalities, all providing data at differing rates, and with varying degrees of uncertainty. These factors complicate the fusion problem as multiple observation models need to be derived, along with a dynamic prediction model. However the problem is exacerbated when the sensors are not correctly registered with respect to one another. Measurements from different sensors may correspond to the same target, but will not correlate with one another when projected into a common reference frame. Recent advances in computationally efficient and scalable tracking and estimation methods allow for both the registration and fusion problems to be solved jointly. This talk will give an overview of the problems, and our approach to solving them, from an application-driven and practical perspective.