Vehicle-to-everything (V2X) in the autonomous vehicles domain – A technical review of communication, sensor, and AI technologies for road user safety

Abstract

Autonomous vehicles (AV) are rapidly becoming integrated into everyday life, with several countries anticipating their inclusion in public transport networks in the coming years. Safety measures in the context of Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) communication have been extensively investigated. However, ensuring safety measures for the Vulnerable Road Users (VRUs) such as pedestrians, cyclists, and e-scooter riders remains an area that requires more focused research effort. The existing AV sensor suites offer diverse capabilities, covering blind spots, longer ranges, and resilience to weather conditions, benefiting the V2V and V2I scenarios. Nevertheless, the predominant emphasis has been on communicating and identifying other vehicles, leveraging advanced communication infrastructure for efficient status information exchange. The identification of VRUs introduces several challenges such as localization difficulties, communication limitations, and a lack of network coverage. This review critically assesses the state-of-the-art in the domains of V2X and AV technologies, aiming to enhance the identification, tracking, and localization of VRUs. Additionally, it proposes an end-to-end autonomous vehicle motion control architecture based on a temporal deep learning algorithm. The algorithm incorporates the dynamic behaviors of both visible and non-line-of-sight (NLOS) road users. The work also provides a critical evaluation of various AI technologies to improve the VRU message sharing, identification, tracking and communication domains.