Crossing Decision

Crossing decisions, encompassing pedestrian and autonomous vehicle navigation at intersections, are a critical area of research aiming to improve safety and efficiency. Current studies employ diverse approaches, including reinforcement learning models that incorporate human perceptual limitations and machine learning algorithms (like random forests and neural networks) to predict crossing behavior based on environmental factors and individual characteristics. These advancements are crucial for enhancing the safety of both human and autonomous systems by providing more accurate predictions of crossing behavior and informing the design of safer infrastructure and control systems.