CPP-OTPO: A Cognitive Multi-UAV System for Adaptive Path and Trajectory Optimization

Unmanned Aerial Vehicles (UAVs) have significantly transformed various Search and Rescue (SAR) operations. Effective resource deployment is still difficult in modern SAR missions, especially in complex and unexpected circumstances. Conventional techniques for multi-UAV systems for SAR operations frequently have difficulties responding to changing conditions. This research proposes a unique method of combining 6G communication technology and Deep Reinforcement Learning (DRL) to improve the cognitive capacities of multi-UAV systems for cooperative coverage to enhance autonomous decision-making in challenging environments. The proposed Coverage Path Planning-Online Time-based Policy Optimization (CPP-OTPO) system utilizes DRL to imbue a multi-UAV framework capable of enabling adaptive decision-making based on realtime observations. The integration of 6G communication further enhances connectivity among multi-UAV systems. The system incorporates boundary search techniques in UAV path planning to achieve optimized CPP, emphasizing resilience against navigational uncertainties. Additionally, the proposed OTPO improves the effectiveness of the target search and increases the possibility of detecting targets by guiding UAVs to change their search policies from exploiting to exploring when required. Empirical evaluations demonstrate a remarkable 92.2% operational time efficiency compared to traditional algorithms.