A Hybrid Priority-Laxity-Based Scheduling Algorithm for Real-Time Aperiodic Tasks Under Varying Environmental Conditions

Efficient task scheduling is fundamental to real-time systems, where minimizing deadline misses and improving stability and determinism are the main goals. In this paper, we propose an algorithm for dynamic task scheduling specifically designed for real-time workloads on homogenous platforms by integrating hybrid laxity-priority scheduling to manage aperiodic tasks with varying priorities. The design includes limiting preemption overhead by blocking unnecessary preemptions and dynamically adapting to incoming sensor data. We extend our algorithm by taking environmental conditions into account, where the weight of laxity and priorities can change based on the circumstances. This also includes the effect of external conditions, namely weather conditions on the priority of tasks. The scheduler is designed for systems such as ADAS, where tasks have different priorities, but are not necessarily dependent on each other. Aperiodic tasks are harder to schedule due to the many unpredictable factors that make up the task. Focusing on aperiodic tasks, the proposed multi-core task scheduler design handles sensor-triggered events and adapts scheduling dynamically by prioritizing running tasks by blocking unnecessary preemption. Experimental results using software-based validation demonstrate improved deadline adherence, affirming the performance of the proposed scheduler compared to classical and real-time schedulers. This study evaluates the scheduler’s performance based on key metrics including deadline misses, preempting, makespan, and the proportion of high-priority deadline misses relative to total deadline misses. Our proposed algorithm demonstrated a miss rate 6% lower than existing state-of-the-art algorithms for similar tasksets.