Wireless Energy Transfer for UAV (Drone) Using Machine Learning

The merging of unmanned aerial vehicles (UAVs) with wireless energy transfer (WET) using deep reinforcement learning is innovative. To conflict UAVs’ limited flight endurance due to small batteries, this system employs machine learning to optimize energy transfer during flight. Real-time data on flight, environment, and energy consumption guide a reinforcement learning it trained UAV to adjust transmission parameters like frequency and power level for efficient energy collection. By continuously learning from feedback, the UAV adapts to changing conditions, enhancing energy transfer efficiency. While prior work achieved a 57% energy collection rate using deep deterministic policy gradient (DDPG), this project aims to exceed it by implementing cutting-edge reinforcement learning algorithms. This innovative approach not only tackles battery limitations but also holds promise for diverse industries like surveillance, disaster management, and agriculture, revolutionizing UAV capabilities for prolonged, high-performance operations.