In-Fleet Connected and Autonomous Vehicle Data-as-a-Service (CAV-DaaS) for Intelligent Transportation Systems and Traffic Control

This book introduces a novel intelligent transportation system (ITS) that formulates the concept of connected autonomous vehicle data-as-a-service (CAV-DaaS). By leveraging crowdsourced data from CAVs, the platform enables autonomous, real-time, continuous, and network-wide monitoring of roadway infrastructure, bridge structural health, traffic conditions, and on-road hazards. Built on a digital twin-enabled system-of-systems architecture, this system provides a scalable alternative to conventional fixed and mobile sensor-based monitoring systems, supporting more efficient, resilient, and intelligent transportation infrastructure management. This methodology crowdsources data from CAV fleets already circulating on the network, leveraging GPS, IMUs, LiDAR, cameras, radar, and vehicle-to-everything links to enable a continuous and autonomous monitoring system that is scalable and without costs attributed to fixed sensors or dedicated test vehicles. The book demonstrates how CAV data can be used for detection and assessment of on-road debris (e.g., tire/cargo debris, rocks and gravel, detached vehicle parts), and road surface damages (e.g., potholes, cracks), road asset monitoring (e.g., signs, road markings, traffic lights), structural health monitoring of roadway bridges (detection of damage and changes in bridge dynamic response), as well as inclusive traffic management and control (e.g., traffic flow visualization, traffic prediction, traffic forecasting, incident detection and management, parking management). The book guides readers from foundational CAV concepts such as autonomy-enabling sensors, perception pipelines, and edge/fog/cloud computing architectures to the advanced algorithms that transform raw vehicular observations into actionable intelligence. The digital twin-enabled platform is structured into four logical layers, with the vehicle perception layer harvesting multimodal sensor data, the communication layer transporting and pre-processing these streams, a cloud-based digital-twin layer fusing, storing, and analyzing the data, and a decision-interface layer returning insight directly to vehicles, roadside controllers, or traffic-management centers.