Identifying Driver Interactions via Conditional Behavior Prediction
This work addresses a key challenge in autonomous driving by improving behavior prediction in interactive scenarios, though it is incremental as it builds on existing prediction methods with a new scoring approach.
The paper tackled the problem of modeling interactive driving scenarios for autonomous vehicles by developing end-to-end conditional behavior prediction models that predict other agents' trajectories based on a query ego-agent trajectory, and introduced an interactivity score for scenario selection and agent prioritization, showing effectiveness in computational budget constraints.
Interactive driving scenarios, such as lane changes, merges and unprotected turns, are some of the most challenging situations for autonomous driving. Planning in interactive scenarios requires accurately modeling the reactions of other agents to different future actions of the ego agent. We develop end-to-end models for conditional behavior prediction (CBP) that take as an input a query future trajectory for an ego-agent, and predict distributions over future trajectories for other agents conditioned on the query. Leveraging such a model, we develop a general-purpose agent interactivity score derived from probabilistic first principles. The interactivity score allows us to find interesting interactive scenarios for training and evaluating behavior prediction models. We further demonstrate that the proposed score is effective for agent prioritization under computational budget constraints.