Multi-Objective Path Planning for UAV Swarm-Assisted Wireless Information and Power Transfer System Based on Decentralized MADRL

With the rapid development of the Internet of Things (IoT), unmanned aerial vehicle (UAV) has become a key component of IoT networks, capable of providing simultaneous wireless information and power transfer (SWIPT) services for widely distributed ground IoT devices. In this paper, we design a UAV swarm-assisted SWIPT system and propose a multi-objective path planning (MOPP) problem to comprehensively consider the sum of data rate, total harvested energy, UAVs’ energy consumption, service fairness and collision avoidance. In addition, owing to the rapid alterations in communication channels resulting from the high mobility of UAVs, centralized management schemes exhibit inherent limitations. To solve this problem, we extend the soft actor-critic (SAC) and propose an enhanced independent SAC (EI-SAC) algorithm based on decentralized multi-agent deep reinforcement learning (MADRL) to address the path planning and resource allocation of UAV swarm. The simulation results show that the proposed EI-SAC algorithm demonstrates superior efficiency and achieves better final performance than other traditional methods.