The Motion Message Estimator in Networked Control Systems

Abstract The networked control system (NCS) in industrial applications, such as computerized numerical control (CNC) machine tools and robots, possess advantages of simple wiring, expansion flexibility, and easy maintenance. However, the introduced network delay is unavoidable. Moreover, in real-time synchronized multi-axis NCS, all motion messages transmitted through the network are required to be received in time to meet specifications of the deadline. However, the data dropout may occur in NCS due to its time delay in a stochastic nature, and it also may results in degradation of motion accuracy. In this paper, a motion message estimator is proposed to construct the real-time networked industrial applications to reduce the data-dropout effect. Analytical results indicate that a 3 rd -order message estimator based on the Taylor expansion successfully suppresses the uncertainties in NCS due to the dropout effect. Furthermore, this paper also proposes the networked cross-coupled control (CCC) structure to improve the contouring accuracy of the multi-axis NCS. Results indicate that the variation of the contouring error in the multi-axis NCS is suppressed greatly by applying the proposed networked CCC. Finally, experimental results indicate the proposed motion message estimator leads to improved accuracy on an industrial CNC machine tool with the NCS implementation.