Research On Feedback Scheduling Of Real-Time Control Systems With Resource Constraints

With rapid evolution of information technologies, real-time control systems (RTCS) are becoming ever-increasingly resource limited. At the same time, due to the dynamic changes of many uncertainties and running load, real-time control systems always have to run in unexpected open environments. In this new implementation environment, traditional control systems design and implementation methodology, which separates control from scheduling, cannot always provide quality of control (QoC) guarantees.This thesis observes the emerging trend of convergence of control with computing. Following the methodology of control and scheduling codesign, some problems in RTCS from a unique viewpoint of dynamic resource constraints are examined. A framework of integrated control and scheduling codesign to those problems is established. And a number of advanced control algorithms based on the real-time feedback control system scheduling methods are made. Considering the CPU time and energy which are two representative kinds of resource constraints, a feedback scheduling technique of computing resources based on support vector machines and a feedback scheduling technique based on fuzzy logic for power-aware systems are designed respectively. Main contents are as follows:As regarding CPU scheduling, we are concerned with multitasking embedded control systems where the processing capacity of the processor is limited. Due to the dynamic changes of many uncertainties and running load, the system always runs in unexpected open environments. The feedback scheduler of the system, which monitors the system's computing resources periodically, predicts a next sampling period for each control task by using SVM mechanism. Consequently, the computing resources of the system can be dynamically allocated online. The simulation of results highlights that the proposed scheduling technique is effectivemethod in comparison with ideal case and traditional open-loop scheduling technique.In the part of energy management, we aim to reduce the energy consumption of the processor as much as possible while preserving the QoC of embedded control systems. Using energy consumption as low as possible to obtain the overall system's quality of control is always an objective of energy management in embedded control system. Due to the dynamic changes of some uncertainties and workload, the embedded systems usually run in unexpected open environments. In order to effectively deal with the change of CPU workload and uncertain execution time of the control tasks, a feedback scheduling technique based on fuzzy logic for power-aware systems is designed. The feedback scheduler, which adjusts the CPU processing speed dynamically and controls the CPU utilization effectively, provides an effective method for energy management mechanism. The simulation results highlight that the proposed scheduling technique is an effective method in comparison with general dynamic voltage scaling (DVS) technique, traditional scheduling technique and ideal case.