Title :
Cooperative source localisation of multi-agent system based on a cooperative PE condition
Author :
Jing-Ping Shao ; Yu-Ping Tian
Author_Institution :
Sch. of Autom., Southeast Univ., Nanjing, China
Abstract :
This study deals with cooperative source localisation with a group of agents where the source is static/dynamic. In stead of the traditionally defined persistent excitation (PE) condition, it is shown that a looser condition - cooperative PE condition - is sufficient for the source localisation provided the topology graph is connected. An estimator is designed to estimate the relative coordinate among agents; a local cooperative estimator based on cooperative PE condition is designed to estimate the relative location of the source; based on the first two steps, a consensus-based estimation fusion algorithm is given to guarantee that estimation errors asymptotically converge to 0. The cooperative localisation method is then extended to the case of a dynamic source and is analysed for the disturbance case. Finally, several simulation results are provided to show the correctness of the conclusion and the effectiveness of the proposed method.
Keywords :
graph theory; multi-robot systems; consensus-based estimation fusion algorithm; cooperative PE condition; cooperative localisation method; cooperative source localisation; multi-agent system; persistent excitation condition; topology graph;
Journal_Title :
Control Theory & Applications, IET
DOI :
10.1049/iet-cta.2014.0049