Multi-Agent Integrated Path Planning and Scheduling in Advanced Air Mobility
Dhananjay Tiwari, Salar Basiri, Srinivasa M. Salapaka · 2025
Advanced Air Mobility (AAM) envisions enhancing urban airspace transportation by improving commute times and energy efficiency. In this paper, we propose a framework for simultaneous scheduling and routing in multi-agent scenarios within AAM. Our approach is grounded in the Facility Location and Path Optimization (FLPO) framework, in which the goal is to identify facility locations that encompass a broad set of underlying nodes and devise shortest transportation paths from each node to a destination center. We model the routes and schedules of the agents as independent FLPO frameworks, and formulate a central mixed integer programming (MIP) problem that minimizes the individual transportation costs. Further, a conflict avoidance cost is added in the MIP to design the conflict-free schedules and routes. This problem is $\mcal N \mcal P$-hard and highly non-convex. To overcome these challenges, we employ a Maximum Entropy Principle (MEP) based solution approach, which is efficient in avoiding many poor local minima. Our results show the effectiveness of this approach in optimizing schedules and routes for multi-agent systems.