Tackling Existence Probabilities of Objects with Motion Planning for Automated Urban Driving
This addresses the operational reliability issue for automated driving systems by reducing the impact of perception and prediction faults, though it appears incremental as it builds on existing motion planning frameworks.
The paper tackles the problem of overcautious behavior in automated urban driving motion planners caused by false positive object detections by proposing a planner that considers alternative maneuvers based on object existence probabilities, resulting in smoother reactions to low-probability objects while maintaining collision-free safety.
Motion planners take uncertain information about the environment as an input. The environment information is often quite noisy and has a tendency to contain false positive object detection. State-of-the-art motion planners consider all objects alike, thus producing overcautious behavior. In this paper we present a planning approach that considers alternative maneuvers in a combined fashion and plans a motion that is formed by the probabilities of those alternatives. The proposed planner can smoothly react to objects with low existence probability while remaining collision-free in case their existence substantiates. In this way, it tolerates the faults arising from perception and prediction, thus reducing their impact on operational reliability.