Deep Reinforcement Learning Based Freshness-Aware Path Planning for UAV-Assisted Edge Computing Networks with Device Mobility

As unmanned aerial vehicles (UAVs) can provide flexible and efficient services concerning the sparse network distribution, we study a UAV-assisted mobile edge computing (MEC) network. To satisfy freshness requirement of IoT applications, age information (AoI) is incorporated as an important performance metric. Then, path planning problem formulated to simultaneously minimize AoIs devices energy consumption UAV, where movement randomness are taken into account. Concerning dimension explosion, deep reinforcement learning (DRL) framework exploited, double Q-learning (DDQN) algorithm proposed realize intelligent freshness-aware UAV. Extensive simulation results validate effectiveness scheme unveil effects moving speed UAV on achieved AoI.