HYPE: Hybrid Planning with Ego Proposal-Conditioned Predictions
This addresses the problem of safe and interpretable motion planning for autonomous vehicles in urban settings, representing an incremental improvement by simplifying cost function design.
The paper tackles the challenge of designing cost functions for motion planning in complex urban environments by proposing HYPE, a hybrid planner that integrates learned trajectory proposals into Monte Carlo Tree Search, achieving state-of-the-art performance in safety and adaptability on benchmarks like nuPlan and DeepUrban.
Safe and interpretable motion planning in complex urban environments needs to reason about bidirectional multi-agent interactions. This reasoning requires to estimate the costs of potential ego driving maneuvers. Many existing planners generate initial trajectories with sampling-based methods and refine them by optimizing on learned predictions of future environment states, which requires a cost function that encodes the desired vehicle behavior. Designing such a cost function can be very challenging, especially if a wide range of complex urban scenarios has to be considered. We propose HYPE: HYbrid Planning with Ego proposal-conditioned predictions, a planner that integrates multimodal trajectory proposals from a learned proposal model as heuristic priors into a Monte Carlo Tree Search (MCTS) refinement. To model bidirectional interactions, we introduce an ego-conditioned occupancy prediction model, enabling consistent, scene-aware reasoning. Our design significantly simplifies cost function design in refinement by considering proposal-driven guidance, requiring only minimalistic grid-based cost terms. Evaluations on large-scale real-world benchmarks nuPlan and DeepUrban show that HYPE effectively achieves state-of-the-art performance, especially in safety and adaptability.