LAUSR

The vast development of the next-generation network (NGN) impels its integration with emerging technologies, such as big data, artificial intelligence, and federated learning, to deliver autonomous and intelligent services in various areas. Notably, in modern transportation systems (TSs), the advances of NGN enable a transformation toward an autonomous transportation system (ATS), which can bridge the demand and supply through a self-actuating cycle (sensing, learning, rearranging, and reacting). Since NGN-enabled ATS is still in its infancy, a concrete vision is missing to forge a common research ground. To fill the gap, this article is intended to elucidate NGN-enabled ATS by first discussing its intrinsic difference against the conventional TSs (CTSs) and then depicting its service blueprint in fostering more intelligent and autonomous mobility services. After that, a full-scale ATS service design reference is proposed to ensure the generality, adaptivity, compatibility, interoperability, and scal-ability of services in and across its development stages, representing the levels of autonomy from partial to high to full automation. Furthermore, its superiority is discussed through a preliminary evaluation of personal mobility service based on centralized and federated learning. Finally, open questions and future research directions of this emerging topic are also discussed.