一种基于自适应控制的软实时调度算法研究

控制理论应用于软实时系统的调度问题，仅采用经典反馈控制技术存在较大的局限性。针对CPU资源实时调度模型的时变特点，提出一种基于自适应反馈控制结构的软实时调度算法，通过在线实时辨识模型的时变参数，自动校正控制器参数，使控制系统运行在期望的闭环极点，改善了实时系统的动态响应特性，同时对系统模型的参数时变具有较好的鲁棒性，提高了实时系统的稳定性和资源的利用率。

Research on a Soft Real-time Scheduling Algorithm Based on the Adaptive Control Structure

It is heavily limited that only the conventional control approaches are mainly adopted to resolve soft real-time system scheduling problems. Aiming at the time-varying characteristic of CPU resource real-time scheduling model, this paper presents a soft real-time scheduling algorithm based on the adaptive feedback control structure. This algorithm identifies the time-varying parameters of the model by online identification. The controller parameters are revised automatically. The control system runs at the expected pole of the closed loop. The robust is demonstrated when the system parameter is varied,and its steady and dynamic response performances are improved. The utilization of resource is increased.