Cooperative Driving and Collective Navigation Architectures for CCAVs: Contemporary Techniques, Simulators, Datasets, and Research Directions

As vehicle-to-vehicle and vehicle-to-infrastructure technologies evolve, connected and cooperative autonomous vehicles (CCAVs) become a feasible solution for traffic problems with their shared and collective navigation and collaborative driving policies to globally optimize safety and efficiency. This chapter covers the architectures in autonomous driving and autonomous navigation of CCAVs. Multi-agent reinforcement learning-based autonomous driving techniques and deep-learning-based end-to-end cooperative driving solutions are investigated. Intermediate fusion schemes for autonomous navigation, and in particular transformer-based collective navigation architectures of CCAVs, are examined. Popular simulators and datasets used in the development and benchmarking of the autonomous driving and navigation architectures of CCAVs are analyzed. Future research directions are discussed.