Governing fully autonomous vehicle adoption: A state-of-the-art review of public acceptance, system impacts, and policy pathways

Source abstract

Fully autonomous vehicles (AVs, Level 4+) represent a transformative transportation technology with the potential to improve traffic efficiency and road safety by enabling optimized traffic operations without human intervention. Their adoption, however, is expected to generate cascading effects on infrastructure provision, mobility services, and urban development. While extensive research has investigated the technical and infrastructural dimensions of AVs, comparatively limited attention has been given to their behavioral and operational dimensions, particularly public acceptance and the system-level implications of AV adoption for planning and policy. This paper presents a state-of-the-art review of theoretical approaches to AV public acceptance and system-level impact assessment, aiming to advance understanding of perception-to-adoption mechanisms and to identify key directions for future research and governance. The review shows that while dominant frameworks identify correlates of AV adoption intentions, their reliance on cross-sectional data constrains interpretability, particularly with respect to assessing the effectiveness of policy interventions. From an adoption perspective, we apply the traditional four-step travel demand model as an analytical lens to categorize modeling assumptions underlying divergent findings and to highlight overlooked system-level impacts of AV adoption across different modeling steps. Existing AV impact assessments demonstrate substantial uncertainty, reflected in wide variation in projected market shares, demand elasticities, and system-level outcomes, largely due to inconsistent assumptions about adoption contexts, automation levels, and infrastructure readiness. To address these limitations, this review proposes an integrated human-vehicle-infrastructure framework that embeds user perceptions, public acceptance, AV adoption dynamics, and transportation system impacts within a feedback-based structure. By grounding key modeling assumptions in adoption outcomes generated through this feedback loop, the framework helps bridge the gap between user-centered acceptance research and system-level impact analysis. The framework further emphasizes the moderating roles of automation level and infrastructure readiness, and the critical need for empirical evidence from real-world AV testing, as well as action priorities and policy pathways among involved stakeholders. These insights jointly offer a coherent and policy-relevant foundation for the integration of fully autonomous vehicles into evolving transportation systems.