Beyond Waypoint Regression: Query-Based Cost Learning over Reachable Ego Futures for End-to-End Driving

2026-10-06

AI Summary

This study addresses the challenge that end-to-end planners struggle to adapt to dynamic safety constraints during deployment. To overcome this, it proposes a query-based cost learning framework that abandons dense BEV representations and fixed trajectory sets, instead estimating bounded costs via dynamically reachable trajectory queries. By integrating emergency-aware cost aggregation, scene tokens, and multimodal agent prediction, the method generates safe and feasible plans using a cost-guided MPPI hybrid policy. Evaluations on the nuScenes dataset demonstrate that this approach surpasses baselines such as ST-P3, significantly reducing collision rates while maintaining competitive L2 error. Furthermore, the framework enhances safety on real-world driving logs without requiring fine-tuning.

Abstract

End-to-end planners based on waypoint regression achieve strong open-loop accuracy, but they primarily learn to mimic expert geometry and remain difficult to adapt to deployment-time safety constraints. We propose a query-based cost-learning framework that estimates bounded costs for dynamically reachable ego trajectory queries, rather than dense BEV cells or a small regressed trajectory set. Compact joint scene tokens capture coherent multimodal agent futures, while contingency-aware cost aggregation and cost-guided intra-cluster MPPI mixing convert the learned cost topology into feasible ego plans. On nuScenes, our method improves over prior cost-estimation planners such as ST-P3 and NMP, outperforms most regression baselines in collision rate, while remaining competitive in L2, and retaining an interpretable cost interface. On real-world driving logs, the proposed planner reduces collision rates compared with SparseDrive and Alpamayo without fine-tuning, while maintaining a diverse set of candidate trajectories.

Problem

• end-to-end driving
• waypoint regression
• safety constraints
• cost learning

Innovation

• Query-based cost learning
• Reachable ego futures
• Contingency-aware cost aggregation
• Cost-guided MPPI mixing
• End-to-end driving