Intelligent Perception-Based Human-Vehicle-Road Collaborative Travel Information Service System

Intelligent Vehicle Infrastructure Cooperative Systems (IVICS) represent cutting-edge technology and an inevitable trend in the global intelligent transportation domain. They serve as an effective means to enhance safety, improve efficiency, optimize energy consumption, and reduce emissions. IVICS elevate traditional road traffic flow theories-based on aggregate models-to a new paradigm grounded in precise descriptions of traffic entities. This paper positions IVICS as a fundamental public platform for future road transportation systems, with a focus on analyzing the resulting evolution in service domain classifications within the National Intelligent Transportation Systems (ITS) architecture framework. On this basis, the paper introduces critical technologies required for constructing and applying IVICS. Key system-building technologies encompass multi-modal communication, intelligent connectivity, information security, and system integration, while core application technologies include collaborative perception, collaborative decision-making and control, simulation testing and validation, and autonomous driving. Considering the long-term nature of IVICS implementation, the paper delineates primary development phases, corresponding levels of information sharing, and achievable collaborative functionalities. Addressing challenges in China’s IVICS deployment, the study emphasizes that deepening the understanding of IVICS technical principles, grasping their operational essence, enhancing user service experiences, and appropriately accelerating large-scale adoption can effectively advance modern intelligent transportation systems. Advancing fundamental theoretical research, key technological development, and practical system applications of IVICS is pivotal to the future construction of intelligent transportation systems and development of related disciplines. The Human-Vehicle-Road Collaborative System (HVRCS) proposed in this study enhances the interaction between vehicles, roads, and the surrounding environment by integrating multi-source data through collaboration, overcoming the limitations of traditional sensing technologies. To further strengthen the system’s driving safety assurance capabilities, this research introduces a novel multifeature fusion method in the fatigue driving monitoring system, which integrates facial features obtained from Dlib and OpenCV. Experimental results demonstrate that the proposed method significantly improves the accuracy of fatigue driving detection. Additionally, the vehicle-road collaborative system has been shown to effectively enhance urban traffic conditions.