Autonomous vehicle groups

The advancement of autonomous technology has elevated the significance of autonomous vehicles in contemporary transportation systems. Currently, individual autonomous vehicles exhibit limited perception range and insufficient autonomous driving capabilities. Cooperative groups of autonomous vehicles, facilitated by advanced communication technologies, enhance traffic efficiency through information exchange. Existing research predominantly focuses on centralized autonomous vehicle groups, where the leading node faces challenges of weak resilience and heavy computational load, posing difficulties in maintaining group cohesion over time. To address these issues, this paper introduces a decentralized formation and self-adaptation method for autonomous vehicle groups based on consensus resistance in closed scenes. First, we propose consensus resistance to assess consistency of group and its member, and devise a decentralized formation approach for autonomous vehicle groups. Subsequently, we present an self-adaption model for autonomous vehicle groups, incorporating four evolutionary events: expansion, merging, reduction, and splitting, to maintain the viability of moving vehicle groups. Simulation results demonstrate the proposed method can effectively construct vehicle groups in both real-world and simulated environments, showcasing robust consistency throughout the self-adaption process.