Multiagent Deep Reinforcement Learning for Wireless-Powered UAV Networks

Unmanned aerial vehicles (UAVs) have attracted much attention lately and are being used in a multitude of applications. But the duration sky remains to be an issue due their energy limitation. In particular, this represents major challenge when UAVs as base stations (BSs) complement wireless network. Therefore, execute missions sky, it becomes beneficial wirelessly harvest from external adjustable flying sources (FESs) power onboard batteries avoid disrupting trajectories. For purpose, transfer (WPT) is seen promising charging technology keep flight allow them complete missions. work, we leverage multiagent deep reinforcement learning (MADRL) method optimize task between FESs UAVs. The optimization performed by carrying out three essential tasks: 1) maximizing sum-energy received all based on using WPT; 2) optimizing loading process ground BS; 3) computing most energy-efficient trajectories while duties. Furthermore, ensure high-level reliability transmission, use directional for both laser beams beam-forming technologies, respectively. study, simulation results show that proposed MADRL has efficiently optimized consumption FESs, which translates into significant gain compared baseline strategies.