On-line Distributed Bayesian decision and diagnosis of Wireless Networked mobile robots

Abstract

This paper deals with the problem of co-design of Wireless Networked Control systems (WNCS). The integration of the Wireless Network (WN) in the control loop influences the Quality of Control (QoC) of the system in terms of the Quality of Service (QoS) of the WN. In this paper, a new model of the QoS and its influence on the QoC is proposed. This model is based on a distributed Bayesian Network (BN). The main objective of this model is to diagnose on-line problems related to the QoS and the QoC. According to the diagnosis results and the situation, the system makes on-line decision to reconfigure both the WN in a multi-station environment and the controller in order to avoid the degradation of the QoC. This reconfiguration is justified by the need of adapting the WN to the requirements of the control and diagnosis, and also adapting the control and diagnosis to the capabilities of the WN. The proposed reconfiguration permits to guarantee a sufficient QoC under network limits.