Optimal on-line (m,k)-firm constraint assignment for real-time control tasks based on plant state information

In this paper, we study the problem of scheduling a set of control tasks. We distinguish three different situations of states of controlled plants: not activated, steady state situation and transient situation. The infinite-horizon and finite-horizon cost functions are respectively used to represent the performance of each control task in last two situations. We propose a scheduling architecture in which, according to the plant state situation, the task handler switches between these two types of performance criterion to determine an on-line (m,k)-constraint based control task scheduling strategy, so that the overall control performance is maintained at a high level in each situation subject to the task schedulability. The approach is exemplified on a set of controllers for different plants.