Abstract: Rural areas present unique challenges for the deployment of autonomous vehicles (AVs), as their physical infrastructures often lack the consistency and quality needed to support reliable automated operation. This research aims to accelerate AV integration by systematically assessing the readiness of rural road infrastructure and developing enhancement strategies to improve compatibility. The study employs a combination of field testing and data-driven analysis using commercial low-level AVs equipped with advanced driver assistance systems (ADAS). These vehicles, equipped with cameras and GPS, collect data on lane markings, signage, and network coverage across selected sites in Georgia. 

A comprehensive software tool is developed to process multimodal sensor data using advanced computer vision and machine learning algorithms. Specifically, the tool integrates YOLOv12 for real-time detection and classification of key roadway elements such as lane markings and traffic signs, while Perspective-n-Point (PnP) techniques is employed to estimate three-dimensional positions of these features from camera data. This fusion of object detection and geometric localization enables precise, quantitative evaluations of infrastructure readiness under diverse environmental conditions. The study further develops a readiness index for infrastructure evaluation, encompassing factors such as sign coverage, detection range, and communication delay. Based on this assessment, it proposes a comprehensive set of infrastructure enhancement strategies—including expansion, maintenance, and rehabilitation—to support data-driven decision-making. Ultimately, the study offers cost-effective solutions to strengthen rural roadways for autonomous vehicle deployment, ensuring that the benefits of AV technology—such as improved safety, mobility, and efficiency—extend beyond urban centers.