Elastic Periodic Task Model Based on Feedback Control

An increasing number of real-time applications require greater flexibility than classical real-time theory usually permits. In this paper, we present a feedback schedule model and corresponding algorithm for control task schedule, the proposed scheme allocates available CPU resources efficiently among the control tasks by self-adaptively adjusting task sample periods online, high-level CPU utilization can be achieved with a low probability of deadline-missing tasks. Even in the case of system overload, this scheme can still keep the system stable through reducing the service quality of control task properly. The experimental result shows that the elastic periodic task model of feedback control has better control performance compared with the traditional periodic task model