评估即一切：多模态自动驾驶
Evaluation Is All You Need for Multi-Modal Autonomous Driving
浏览论文内容
中文总结 AI 辅助
提出iDriveVLA多模态规划框架，通过统一轨迹评估器与渐进训练策略解决生成-评估不对称，在NAVSIM v1上达到94.95 PDMS，超越人类专家。
中文摘要 AI 辅助
多模态规划在自动驾驶中具有前景，通过在模糊和长尾场景中表示多种合理行为。现有方法主要集中于改进轨迹多模态性、增强轨迹表示或重塑候选分布。然而，我们识别出多模态规划中一个显著的生成-评估不对称性：尽管具有强大的oracle性能，现有规划器往往无法可靠地选择最佳可用候选，留下大量未实现的规划潜力。为解决这一挑战，我们提出iDriveVLA，一种多模态规划框架，改进候选轨迹空间，同时实现更可靠和上下文感知的轨迹评估。具体而言，iDriveVLA引入统一轨迹评估器，包含用于质量和风险估计的安全感知评分器，以及用于场景自适应标准加权的VLM引导调制器。我们进一步开发了oracle对齐的渐进训练策略，包括候选模仿预训练、候选空间细化和语义排名对齐。在公共NAVSIM v1排行榜上，iDriveVLA达到94.95 PDMS的新最先进性能，超越人类专家参考。
英文摘要
Multi-modal planning is promising for autonomous driving by representing multiple plausible behaviors in ambiguous and long-tail scenarios. Existing methods mainly focus on improving trajectory multi-modality, enhancing trajectory representations, or reshaping the candidate distribution. Nevertheless, we identify a pronounced generation-evaluation asymmetry in multi-modal planning: despite strong oracle performance, existing planners often fail to reliably select the best available candidate, leaving substantial planning potential unrealized. To address this challenge, we propose iDriveVLA, a multi-modal planning framework that improves the candidate trajectory space while enabling more reliable and context-aware trajectory evaluation. Specifically, iDriveVLA introduces a unified trajectory evaluator comprising a Safety-aware Scorer for quality and risk estimation, together with a VLM-guided Modulator for scene-adaptive criterion weighting. We further develop an oracle-aligned progressive training strategy consisting of candidate imitation pretraining, candidate space refinement, and semantic ranking alignment. On the public NAVSIM v1 leaderboard, iDriveVLA achieves a new state-of-the-art performance of 94.95 PDMS, surpassing the human-expert reference.
发表机构
- School of Vehicle and Mobility & College of AI, Tsinghua University（清华大学车辆与运载学院与人工智能学院）
- Dongfeng Motor Corporation Research & Development Institute（东风汽车集团有限公司研发总院）
机构由 AI 辅助整理，请以论文原文为准。
