Path Planning for Cellular-Connected UAV: A DRL Solution with Quantum-Inspired Experience Replay

• DOI: 10.48550/arxiv.2108.13184
• Publication Date: 2021-01-01
• Authors: Yuanjian Li; A.H. Aghvami; Daoyi Dong
• Keywords: Signal Processing (eess.SP); FOS: Electrical engineering, electronic engineering, information engineering; 0202 electrical engineering, electronic engineering, information engineering; Systems and Control (eess.SY); 02 engineering and technology; Electrical Engineering and Systems Science - Signal Processing; Electrical Engineering and Systems Science - Systems and Control

In cellular-connected unmanned aerial vehicle (UAV) network, a minimization problem on the weighted sum of time cost and expected outage duration is considered. Taking advantage UAV's adjustable mobility, an intelligent UAV navigation approach formulated to achieve aforementioned optimization goal. Specifically, after mapping task into Markov decision process (MDP), deep reinforcement learning (DRL) solution with novel quantum-inspired experience replay (QiER) framework proposed help find optimal flying direction within each slot, thus designed trajectory towards destination can be generated. Via relating experienced transition's importance its associated quantum bit (qubit) applying Grover iteration based amplitude amplification technique, DRL-QiER commits better trade-off between sampling priority diversity. Compared several representative baselines, effectiveness supremacy are demonstrated validated in numerical results.