Graph Attention Network-Based Cooperative Trajectory Planning for Multi-UAV Collision Avoidance
Authors:
Open Access
- 6 June 2026
- journal article
- research article
- Published by MDPI AG in Electronics
- Vol. 15 (12) , 2496
Abstract
Trajectory planning for a multi-UAV system requires jointly considering obstacle avoidance, inter-UAV conflict avoidance, and target reaching. To address this problem, this paper proposes a graph attention network-based method for multi-UAV trajectory planning. The multi-UAV system is represented as an interaction graph, where UAVs are modeled as nodes and communication-based inter-UAV relationships are modeled as edges. For each UAV, local perception, target-related direction information, previous motion direction, and neighborhood information are integrated into the node representation, while the relative geometric relationship between neighboring UAVs is used as the edge feature. The constructed graph is fed into a multi-head graph attention network to extract interaction-aware features and output an action score vector over discrete flight direction labels for each UAV. During online execution, candidate flight actions are generated according to the action scores, and the final action is selected using the geodesic cost-to-go map. The trajectories of all UAVs are then generated step by step through the online decision process. By combining local perception, target guidance, motion history, and inter-UAV interaction information, the proposed method can learn cooperative action preferences for multi-UAV trajectory generation. Experiments are conducted on different flight maps and swarm sizes using multiple performance metrics. The results show that the proposed method achieves effective performance in mission success, flight efficiency, and safety-related metrics, and it also demonstrates generalization ability on unseen maps. Compared with a CBF-based collision avoidance method, the proposed method achieves better performance in task completion, inter-UAV collision avoidance, and trajectory efficiency.Keywords
Funding Information
- National Natural Science Foundation of China (62303327)
- Guangdong Provincial Engineering Technology Research Center of Regular Higher Education Institutions (2025GCZX017)
- Programs of Regular Higher Education Institutions by Department of Education of Guangdong Province (2023KTSCX322)
- Science and Technology Program of Shenzhen (20220818100434001)
This publication has 36 references indexed in Scilit:
- Variable admittance control for door opening with a wheeled mobile manipulator considering ground obstaclesIntelligent Service Robotics, 2026
- Dynamic Prioritization and Adaptive Path Planning for Indoor Multi-object NavigationInternational Journal of Control, Automation, and Systems, 2025
- Energy-efficient altitude optimization in multi-UAV search and rescue: A hybrid swarm approachInternet of Things, 2025
- Deadlock-free, safe, and decentralized multi-robot navigation in social mini-games via discrete-time control barrier functionsAutonomous Robots, 2025
- An adaptive heuristic algorithm with a collaborative search framework for multi-UAV inspection planningApplied Soft Computing, 2025
- Graph-based multi-agent reinforcement learning for large-scale UAVs swarm system controlAerospace Science and Technology, 2024
- Cooperative task allocation with simultaneous arrival and resource constraint for multi-UAV using a genetic algorithmExpert Systems with Applications, 2023
- Distributed cooperative search method for multi-UAV with unstable communicationsApplied Soft Computing, 2023
- Flight formation of UAVs in presence of moving obstacles using fast-dynamic mixed integer linear programmingAerospace Science and Technology, 2016
- Trajectory optimization of multiple quad-rotor UAVs in collaborative assembling taskChinese Journal of Aeronautics, 2016