On-line identification of robotic ubiquitous cognitive network with erasure channels

Nowadays the Robotic Ubiquitous Cognitive network (RUBICON) has raised great attentions in robotics area. In this paper we explore and first propose a timely identification method of RUBICON with erasure channels. Firstly, the complex cognitive network is simplified into a single close-loop communicating network for identification. Then inspired by the interactive cognitive method, we propose recursive network identification method containing feedback mechanism. Three key problems of this method - dense format of transmitting messages, selection of prior packet for identification and the detailed network identification procedure - and their solving methods are presented. Finally, we make a simulation about the parameters identification of RUBICON which contains 8 sensors/actuator nodes. The simulated results show that the proposed algorithm can balance the difference of the identification results for every sensor node and enhance the unique convergence effect.