ABSTRACT
Spatiotemporal AI has played an important role in transportation research since the latter part of the 20th century, predating the origin of the term as a way of describing AI based methods that take into account the properties of spatiotemporal data. Transport networks are closely monitored by transport authorities, generating vast amounts of big data that are used in intelligent transportation systems to predict and manage congestion, control signal timings and monitor safety, amongst other things. Spatiotemporal AI methods have facilitated these tasks due to their ability to model the underlying complexity and heterogeneity of transport networks. In this chapter, we review three important application domains of Spatiotemporal AI in transport: 1) data driven prediction of traffic variables; 2) optimization of traffic networks using reinforcement learning; 3) computer vision for sensing complex urban environments. The chapter concludes with some directions for future research in Spatiotemporal AI for transportation.
