Distributed Estimation for Vehicular Collaborative Navigation

Chun Yang, Andrey Soloviev

Abstract:	Exchange of navigation states between networked vehicles (i.e., network nodes) not only supplies the information necessary for situational awareness but also provides a means to obtain relative range, range rate, and/or angular measurements via radio propagation. Individual vehicles (nodes) combine their own a priori navigation state estimate with the a priori estimates of other nodes through relative measurements between them. Each node (a vehicle) in the ad hoc wireless network therefore uses its neighboring nodes (vehicles) in the same network (a single hop) as reference and vice versa, leading to vehicular collaborative navigation with enhanced performance. Different from centralized optimal estimation, vehicular collaborative navigation necessitates distributed estimation carried out at individual nodes because of the limited bandwidth of communication channels that restricts the amount of data that can be exchanged between nodes with a minimum latency for real-time processing imposed by safety applications. Sub-optimality arises for each node can only process partial measurements to estimate a subset of the whole network state. The state of neighboring nodes when used as reference for relative measurements is not perfectly known and, more importantly, the state estimation errors become correlated once the relative measurements between two nodes are processed by the respective estimators. The cross-correlation, if not accounted for properly, would make the estimates too optimistic (the filters thus inconsistent) and reduce the effectiveness of subsequent measurements. This paper illustrates these issues for different network topologies with simulation examples and presents various distributed suboptimal estimation schemes suitable for vehicular collaborative navigation.
Published in:	Proceedings of the 27th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2014) September 8 - 12, 2014 Tampa Convention Center Tampa, Florida
Pages:	211 - 222
Cite this article:	Yang, Chun, Soloviev, Andrey, "Distributed Estimation for Vehicular Collaborative Navigation," Proceedings of the 27th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2014), Tampa, Florida, September 2014, pp. 211-222.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In