Built Road

Built road research focuses on improving our understanding and management of road networks, encompassing aspects from automated road mapping and condition assessment to traffic prediction and autonomous vehicle navigation. Current research employs diverse machine learning models, including graph convolutional networks for traffic prediction, deep learning architectures for pothole detection and road segmentation from various data sources (e.g., satellite imagery, LiDAR, GPS trajectories), and reinforcement learning for optimal road planning in challenging environments like slums. These advancements have significant implications for infrastructure management, transportation efficiency, autonomous driving safety, and urban planning, offering more accurate predictions, improved safety measures, and optimized resource allocation.