Fault-tolerance scheduling for messages and tasks in a class of networked control system

Abstract

Messages and tasks in networked control systems must be executed in certain order and finished before their deadline, and they run on different resources. Therefore, a novel fault-tolerant scheduling algorithm is presented in this paper. Firstly, a scheduling sequence (including messages, primary copies and backup copies) is designed in a sampling period. Then heuristic static algorithm for assigning primary copies and backup copies of tasks to processors is adopted and the primary copy and backup copy of a task are assigned to different processors. The tasks assigned to the same resource are executed according to their orders in scheduling sequence. Because the sampling period and end-to-end delay of control loop influence on the performance of control system, genetic algorithm based on integer coding is applied to optimized scheduling sequence. Simulation result show the scheduling algorithm can guarantee that tasks are fault-tolerant schedulable and improve the performance of control system.