Research on Optimization of Task Scheduling Algorithm for Embedded Real-time Systems

Firstly, this paper summarizes the current research status of the principles of embedded systems and task scheduling strategies, and analyzes the limitations of methods to improve the performance of a single processor. Then, this paper focuses on the task scheduling problem in single-processor embedded systems. By using the time-sharing task scheduler, a commonly used technology in embedded systems, this paper proposes a time-sharing task scheduling algorithm based on adjustable priority. It can dynamically adjust the priority according to the periodicity of different tasks, thereby improving the response speed and execution efficiency of the system. During the design process, we adopted a combined strategy of the round-robin scheduling algorithm and the dynamic priority scheduling algorithm, and experiments proved that this algorithm can improve the performance of the embedded system. In embedded systems with limited resources, the quality of the task scheduler largely determines the performance of the system. This paper analyzes representative static and dynamic scheduling algorithms in real-time systems and summarizes their respective advantages and disadvantages. On this basis, the embedded real-time kernel scheduling algorithm is optimized and applied to vehicle-mounted real-time systems.