Conflict Detection and Resolution Considering Human–Machine and Air–Ground Interaction

With the continual growth of air traffic, controller workload has become a key bottleneck, particularly in high- density airspace where multi-aircraft conflict resolution strains cognitive and computational capacity. While optimisation- and AI-based methods have been proposed, they often assume ad- visories are executed instantly, overlooking delays and uncer- tainties in human–machine and air–ground interactions, which limits their practical applicability. Building on the SESAR HYPERSOLVER project and supporting its vision of advancing human–AI teaming in air traffic management (ATM), this paper proposes a conflict detection and resolution decision-support method that explicitly incorporates controller workload and the dynamics of human–machine and air–ground interactions. This work introduces three key innovations: a trajectory prediction method that captures uncertainty caused by human–machine and air–ground interactions, thereby enabling more realistic conflict detection under dynamic conditions; a robustness index inte- grated into conflict-free trajectory planning optimisation, allow- ing the generation of solutions that better withstand operational uncertainties while ensuring safety and efficiency; and an adap- tive dynamic search algorithm alternating between global and local strategies, effectively balancing optimisation performance and computational efficiency. Fast-time simulations demonstrate that the proposed approach has the potential to effectively handle uncertainties in human–machine and air–ground interactions in real-time operations, while accommodating traffic densities beyond the current maximum observed in practice, thereby paving the way for more effective deployment in real-world ATM systems.