Scene Context-Aware Trajectory Prediction in Shared Space
Led by: | Hao Cheng, Prof. Sester and Prof. Fidler |
Team: | Rui Liu |
Year: | 2019 |
Is Finished: | yes |
In shared spaces, road signs, signals, and markings are removed to allow mixed traffic directly interact with each other. At a micro level, understanding how they behave and how we can foresee their behavior after a short observation time are crucial to intent detection and autonomous driving, and traffic management in shared spaces.
Nevertheless, due to heterogeneity of transport mode, dynamic environment, and various demographic attributes (e.g., age, gender, time pressure) of road users, modeling mixed traffic in shared spaces is of great challenge. In this project, we mainly focus on learning the impact of scene context of shared spaces on road users' behavior. The assumption is that the scene context of a given shared space may constrain certain patterns of behavior, which is crucial for estimating the feasibility of predicted trajectories. For example, vehicles normally have to stay on lanes, and vegetation and curbside may direct pedestrians and cyclists to certain paths. The main questions are: how can we parse the scene context of shared spaces and how can we leverage this information to adapt predictions of trajectories accordingly?