Pedestrian Emergence Estimation and Occlusion-Aware Risk Assessment for Urban Autonomous Driving

Avoiding unseen or partially occluded vulnerable road users (VRUs) is a major\nchallenge for fully autonomous driving in urban scenes. However,\nocclusion-aware risk assessment systems have not been widely studied. Here, we\npropose a pedestrian emergence estimation and occlusion-aware risk assessment\nsystem for urban autonomous driving. First, the proposed system utilizes\navailable contextual information, such as visible cars and pedestrians, to\nestimate pedestrian emergence probabilities in occluded regions. These\nprobabilities are then used in a risk assessment framework, and incorporated\ninto a longitudinal motion controller. The proposed controller is tested\nagainst several baseline controllers that recapitulate some commonly observed\ndriving styles. The simulated test scenarios include randomly placed parked\ncars and pedestrians, most of whom are occluded from the ego vehicle's view and\nemerges randomly. The proposed controller outperformed the baselines in terms\nof safety and comfort measures.\n