Multi-leader Stackelberg Game with Authority-Weighted Exponential Pricing for UAV-Assisted Edge Computing

Unmanned aerial vehicle (UAV) swarms face significant computational constraints when executing complex missions, necessitating efficient computation offloading to edge infrastructure. This paper investigates a multi-UAV multi-edge-device system where geographically clustered edge devices form coalitions to provide spectrum resources through market-based mechanisms. We propose the Geo-Regulated Authority-Incentivized Distributed (GRAID) Stackelberg pricing algorithm, which introduces three key innovations: First, intra-coalition pricing authority is dynamically allocated based on devices’ revenue contributions. Second, an exponential pricing model captures the superlinear relationship between computational capabilities and resource costs. Third, cross-coalition coordination mechanisms prevent predatory pricing through progressive authority penalties and bilateral price stabilization. Comprehensive experimental results confirm that GRAID substantially reduces average task completion time while simultaneously increasing revenue generation for edge devices. The framework provides a robust solution for dynamic edge resource allocation in swarm computing scenarios.