Evaluation of a Power Management Strategy for Autonomous Cyber-Physical Systems with Real-Time Constraints

Designing Energy Harvesting Cyber-Physical Systems (EH CPS) with real-time capabilities requires a sophisticated approach that integrates energy aware scheduling policies, predictive modeling, and dynamic task allocation. By effectively managing energy resources, these systems can adopt an energy neutral behavior: they ensure perpetual operation under real-time requirements even in environments with intermittent energy availability. In response to this challenge, this paper introduces the Extended Earliest Deadline First with Harvesting (EED-H) algorithm, a novel scheduling and power management framework. The EED-H algorithm employs an energy-aware prioritization mechanism, dynamically adjusting job execution based on real-time assessments of available energy, slack energy, and slack time. This adaptive approach not only ensures that jobs meet their deadlines but also optimizes energy utilization during periods of scarcity. By addressing the critical challenges of energy management in CPS, the EED-H algorithm paves the way for more resilient and efficient self-sustaining devices. The experimental results demonstrate that the proposed algorithm reduces the deadline missing rate by 37% when compared to the well known Earliest Deadline First scheduler.