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 2605.14201cs.ROcs.CV
收藏

MAPLE:基于潜在空间的多智能体交互用于端到端自动驾驶

MAPLE: Latent Multi-Agent Play for End-to-End Autonomous Driving

Qualcomm AI Research(英矽人工智能研究)
Rajeev Yasarla, Deepti Hegde, Hsin-Pai Cheng, Shizhong Han, Yunxiao Shi, Meysam Sadeghigooghari, Hanno Ackermann, Litian Liu, Pranav Desai, Fatih Porikli, Moham… 展开作者

Rajeev Yasarla, Deepti Hegde, Hsin-Pai Cheng, Shizhong Han, Yunxiao Shi, Meysam Sadeghigooghari, Hanno Ackermann, Litian Liu, Pranav Desai, Fatih Porikli, Mohammad Ghavamzadeh, Hong Cai

AI总结:

本文提出MAPLE框架,通过在视觉-语言-动作模型的潜在空间中实现反应式多智能体滚动,以解决传统模仿学习框架下闭环设置中模型易碎的问题,通过监督微调和强化学习结合多样性奖励,实现了可扩展且无需外部模拟器的闭环训练,提升了端到端自动驾驶系统的鲁棒性。

摘要:

视觉-语言-动作(VLA)模型在端到端运动规划中表现出色,但在闭环设置中由于训练基于传统模仿学习框架而显得脆弱。现有的闭环监督方法缺乏可扩展性且无法完全建模反应式环境。我们提出MAPLE,一种新的框架,用于在VLA模型的潜在空间中进行动态驾驶场景的反应式多智能体滚动。主体车辆和附近交通代理在多步时间范围内独立控制,同时对场景中的其他代理具有反应性,从而实现闭环训练。MAPLE包含两个训练阶段:(1)基于真实轨迹的潜在滚动监督微调,随后是(2)具有全局和代理特定奖励的强化学习,这些奖励鼓励安全、进展和交互真实感。我们进一步提出多样性奖励,鼓励模型生成可能不在记录驾驶数据中存在的规划行为。值得注意的是,我们的闭环训练框架具有可扩展性,且无需外部模拟器,这些模拟器计算成本高且视觉保真度有限。MAPLE在Bench2Drive上实现了最先进的驾驶性能,并展示了可扩展的闭环多智能体交互,为鲁棒的端到端自动驾驶系统提供了支持。

英文摘要:

Vision-language-action (VLA) models are effective as end-to-end motion planners, but can be brittle when evaluated in closed-loop settings due to being trained under traditional imitation learning framework. Existing closed-loop supervision approaches lack scalability and fail to completely model a reactive environment. We propose MAPLE, a novel framework for reactive, multi-agent rollout of a dynamic driving scenario in the latent space of the VLA model. The ego vehicle and nearby traffic agents are independently controlled over multi-step horizons, while being reactive to other agents in the scene, enabling closed-loop training. MAPLE consists of two training stages: (1) supervised fine-tuning on the latent rollouts based on ground-truth trajectories, followed by (2) reinforcement learning with global and agent -specific rewards that encourage safety, progress, and interaction realism. We further propose diversity rewards that encourage the model to generate planning behaviors that may not be present in logged driving data. Notably, our closed-loop training framework is scalable and does not require external simulators, which can be computationally expensive to run and have limited visual fidelity to the real-world. MAPLE achieves state-of-the-art driving performance on Bench2Drive and demonstrates scalable, closed-loop multi-agent play for robust E2E autonomous driving systems.

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。

↑