A task scheduling scheme for real-time multi-robotics systems

Abstract

In hard real-time systems, the importance of meeting a task execution deadline is a central issue for correctness and reliability of such systems. The focus of this research is to design and implement a real-time system which utilizes a set of mobile robots that are performing individual and common critical tasks. Specifically, a scheduling scheme consists of local and global schedulers is developed to schedule critical tasks using a transputer-based machine. To guarantee execution of critical tasks within the requested time frame, each task is provided with a Safety Time Factor (STF) based on its worst-case computation time. This STF can be added to the execution time of a newly arrived task if its deadline and the status of scheduled tasks allow. The local scheduler takes the advantages of accumulated safety times and provides alternatives for scheduling sporadic critical tasks. The performance of the model is evaluated in terms of the number of critical tasks that are guaranteed and executed by local processors. >