Ubiquitous sensing for smart cities with autonomous vehicles

Advanced autonomous vehicles (AVs)-related technologies stimulate a significant increase in AVs’ sensing capabilities, which accelerates the dramatic development of the smart city, especially in intelligent infrastructure management. Conventional sensing mainly depends on fix-point devices with laborers, which is time consuming, inefficient, and laborious. In contrast, AVs equipped with various sensors, including but not limited to LiDAR, radar, camera, and stereovision, can break the above dilemma in one reliable, ubiquitous, and inexpensive way. However, the current research mainly regards AVs as a participant, controller, or server instead of as an observer. AVs can perform well as floating (or probe) sensors collecting instant, massive, and actual data, which can be utilized to continuously monitor transport infrastructure as well as traffic state during cruising. In this book chapter, we introduce the ubiquitous sensing framework with AVs, fully tapping its tremendous sensing potential to contribute to the smart city. First, levels of sensing capacities, perception, mapping, and localization of AVs are discussed in turn. Second, we summarize and present how AVs play an essential role in various smart city scenarios, specifically in traffic state estimation, on-street parking management, signage identification, pavement maintenance, and pedestrian and safety/health monitoring. Third, both advantages and disadvantages of AVs are unfolded from the perspective of dialectical thinking. Lastly, this chapter ends with some conclusions. Overall, this chapter would help policymakers and private sectors (e.g., Waymo and other AV firms) pay more attention to the ubiquitous sensing capacities of AVs and deeply understand the latent commercial value in empowering smart cities.