发现驾驶数据集隐藏偏差，提升端到端模型表现
Hidden Biases of End-to-End Driving Datasets
- 分析训练数据分布对策略性能的影响，揭示专家风格关键作用
- 通过帧间标签变化判断优化数据权重，压缩数据量同时保留关键信息
- 在CARLA挑战赛中取得榜首成绩，适合自动驾驶数据研究者参考
端到端自动驾驶系统进展迅速，但尚未应用于具有挑战性的CARLA Leaderboard 2.0。尽管已有大量关于架构与训练策略的研究，训练数据的影响常被忽视。本文首次尝试在Leaderboard 2.0上实现端到端驾驶，不聚焦架构改进，而是系统分析训练数据，得出新见解：（1）专家驾驶风格显著影响下游策略性能；（2）复杂数据集中不应仅按类别频率等简单标准加权；（3）通过判断当前帧是否改变目标标签，可有效减少数据规模而不丢失重要信息。基于此，模型在2024 CARLA挑战赛地图与传感器赛道分别排名第一和第二，并在Bench2Drive测试路线创下新基准。最后，我们揭示现有评估指标的设计缺陷并提出改进建议。数据集、代码及预训练模型已公开于https://github.com/autonomousvision/carla_garage。
原文摘要 · Abstract (English)
End-to-end driving systems have made rapid progress, but have so far not been applied to the challenging new CARLA Leaderboard 2.0. Further, while there is a large body of literature on end-to-end architectures and training strategies, the impact of the training dataset is often overlooked. In this work, we make a first attempt at end-to-end driving for Leaderboard 2.0. Instead of investigating architectures, we systematically analyze the training dataset, leading to new insights: (1) Expert style significantly affects downstream policy performance. (2) In complex data sets, the frames should not be weighted on the basis of simplistic criteria such as class frequencies. (3) Instead, estimating whether a frame changes the target labels compared to previous frames can reduce the size of the dataset without removing important information. By incorporating these findings, our model ranks first and second respectively on the map and sensors tracks of the 2024 CARLA Challenge, and sets a new state-of-the-art on the Bench2Drive test routes. Finally, we uncover a design flaw in the current evaluation metrics and propose a modification for future challenges. Our dataset, code, and pre-trained models are publicly available at https://github.com/autonomousvision/carla_garage.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。