Infrastructure Preparedness for Autonomous Vehicles in Rural Areas

While advancements in autonomous vehicle (AV) technology heavily favor urban environments, rural and tribal regions remain severely underprepared for automated deployment due to physical infrastructure deficiencies and communication limitations. To address this disparity, this paper formulates a scalable, low-cost multimodal framework designed to systematically evaluate rural infrastructure readiness across three critical operational domains: communication connectivity, lane marking detectability, and road surface kinematics. First, cellular network reliability is mapped via mobile telemetry to isolate connectivity-deficient risk zones. Second, a vision-based perception module leverages deep learning models on GPS-tagged video footage to quantify lane marking detectability. Third, an onboard road surface assessment pipeline captures inertial measurement unit (IMU) and GPS data to calculate safety-critical geometry metrics-including rollover risk and composite acceleration-while introducing a rapid estimation using the Analytic Hierarchy Process (AHP) of the roadway International Roughness Index (IRI). These sub-indicators are weighted and synthesized into a unified Autonomous Vehicle Readiness Index (ARI) to map to the spatial readiness of AV-critical infrastructure systems. Validated across underserved rural corridors, this framework provides transportation agencies with an objective, data-driven decision engine to prioritize localized infrastructure remediation and accelerate safe, equitable AV deployment.