arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2409.18218cs.ROcs.CVcs.LG
收藏

通过非对称自博弈学习驾驶

Learning to Drive via Asymmetric Self-Play

Waabi(瓦比公司);University of Toronto(多伦多大学)

Chris Zhang, Sourav Biswas, Kelvin Wong, Kion Fallah, Lunjun Zhang, Dian Chen, Sergio Casas, Raquel Urtasun

AI总结:

提出非对称自博弈方法,通过教师-学生配对生成挑战性合成场景,学习真实驾驶策略,减少碰撞并零样本迁移生成端到端自动驾驶训练数据,显著优于现有方法。

摘要:

大规模数据对于学习真实且具备能力的驾驶策略至关重要。然而,仅依赖真实数据来扩展数据集是不切实际的。大多数驾驶数据缺乏趣味性,且刻意收集新的长尾场景既昂贵又不安全。我们提出非对称自博弈,通过额外增加具有挑战性、可解决且真实的合成场景来扩展超越真实数据的规模。我们的方法将一个学习生成自己能解决但学生无法解决的场景的教师,与一个学习解决这些场景的学生配对。当应用于交通模拟时,我们学习到的真实策略在常规和长尾场景中的碰撞次数显著减少。我们的策略进一步实现零样本迁移,为端到端自动驾驶生成训练数据,显著优于最先进的对抗性方法,或仅使用真实数据。欲了解更多信息,请访问 https://waabi.ai/selfplay 。

英文摘要:

Large-scale data is crucial for learning realistic and capable driving policies. However, it can be impractical to rely on scaling datasets with real data alone. The majority of driving data is uninteresting, and deliberately collecting new long-tail scenarios is expensive and unsafe. We propose asymmetric self-play to scale beyond real data with additional challenging, solvable, and realistic synthetic scenarios. Our approach pairs a teacher that learns to generate scenarios it can solve but the student cannot, with a student that learns to solve them. When applied to traffic simulation, we learn realistic policies with significantly fewer collisions in both nominal and long-tail scenarios. Our policies further zero-shot transfer to generate training data for end-to-end autonomy, significantly outperforming state-of-the-art adversarial approaches, or using real data alone. For more information, visit https://waabi.ai/selfplay .

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑