Distributed Kalman Filter algorithms for self-localization of mobile devices

This paper addresses the problem of self localization of mobile devices. In particular, each device combines noisy measurements of its absolute position with distance measurements to its neighbors. The communication topology is modeled by a graph. Both static and dynamic graph structures are investigated. The self-localization task is addressed using distributed Kalman Filters. First a filter is designed which uses only locally available measurements for state estimation. Secondly, a data fusion step is added to the filter. This allows the usage of more measurement information available in the network to improve the accuracy. When the graph is dynamic, a larger communication radius is necessary to ensure reliable performance. Categories and Subject Descriptors

• Type: Conference
• Year: 2010
• Where: HYBRID
• Authors: Anne-Kathrin Hess, Anders Rantzer
• Keywords: Dynamic Graph Structures, HYBRID 2010, Hybrid Systems, Measurement, Noisy Measurements