Automatic Sampling Time Adaptation for Networked Digital Controller based-on Embedded System

Abstract

The sampling time or sampling interval is the importance parameter for all embedded digital controllers. The sampling time of all digital controllers is normally defined as a fixed value, as low as possible. It is commonly known that the lower sampling time (high sampling rate) produces the higher control performance because the system can operate closed to the continuous-time system. However in networked control system, the lower sampling time required higher traffic load or higher bandwidth of the network. If the traffic load exceeds the bandwidth limitation of network, the packet loss and delay-time variation of the network are increased and the system performance will be decreased. In many digital controlled applications, the appropriate sampling time (dt) is need to be defined depended on the plant and limitation of the hardware. This research proposes an automatic adaptive sampling time for embedded networked control system. The sampling time will be continuously adapted to proper value to maintain the controller performance, save computational time, decrease traffic load, and reduce power consumption. The sampling time is changed based on the difference of reference input and output. The simulation results show that the proposed method can produce the output response with control performance’s criteria and can be used in both wire and wireless embedded networked control systems.