Adaptive Task Scheduling Switcher for a Resource-Constrained IoT System

This paper proposes a novel method to use machine learning for switching the scheduling algorithm that has lower computation time and better task execution sequence optimization to meet the computation deadline. Due to the number of tasks and the number of types of resources taken will affect the computation time of the scheduler, the fixed scheduling algorithm unable to meet the computation deadline in worst-case time-complexity. Our implementation of machine learning predicts the best scheduling algorithm to counter the problem, and the result shows it can improve the accuracy of meeting the computation deadline by an average of 85.11%.