Comprehensive Guide to the Industrial Drone Energy Management Dataset

Industrial drones face dual bottlenecks in energy management and decision-making latency. This paper proposes the Smart In-flight Energy (SIF) system, an algorithm-hardware co-design framework integrating lightweight Proximal Policy Optimization (PPO) with spin thermoelectric modules. Core innovations include: embedding the magnetic-thermal gradient (\u2207B\u00b7\u0394T) into a Lyapunov-stable dynamic reward function for real-time policy adaptation, and enhancing thermoelectric conversion efficiency via spin current (Ispin) regulation. Gazebo simulations and DJI M300 field tests demonstrate that the SIF system achieves navigation errors \u22640.48\u00b0 under strong magnetic interference (>100\u03bcT), edge computing latency <80ms, and energy synergy efficiency of 88.6%\u2014a 29.7% improvement over the fuel cell baseline. This work pioneers an AI-new energy collaborative optimization paradigm for industrial drones, increasing daily agricultural coverage by 50% and providing a verifiable technical framework for high-efficiency aerial operations.