Optimizing UAV-UGV Coalition Operations: A Hybrid Clustering and Multi-Agent Reinforcement Learning Approach for Path Planning in Obstructed Environment

The article proposes a novel approach for optimizing the operations of coalitions of Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs). The focus is on efficient path planning in obstructed environments, utilizing a modified mean-shift clustering algorithm and advanced reinforcement learning algorithms. This approach improves the efficiency of reaching targets by segmenting them into multiple zones and assigning targets based on density and range. The method has shown to nearly double efficiency in terms of target navigation time and task completion rate.

Publication date: 5 Jan 2024
Project Page: Not available
Paper: https://arxiv.org/pdf/2401.01481