用概率模型实现可调控的多车交互仿真，提升自动驾驶评估效率。
On Learning Closed-Loop Probabilistic Multi-Agent Simulator
- 基于分层贝叶斯框架，通过潜空间混合高斯分布进行自回归采样。
- 在Waymo数据集上性能媲美现有方法，支持意图与驾驶风格调节。
- 适合自动驾驶研发中需要多样化交互场景的团队使用。
自动驾驶车辆的快速迭代带来对高效、真实且可扩展的多智能体交通仿真器的迫切需求。当前研究聚焦于闭环仿真器，可生成多样且交互式场景。本文提出神经交互代理（NIVA），一种基于分层贝叶斯模型的概率化多智能体仿真框架，通过潜空间有限混合高斯分布的自回归采样实现观察条件下的闭环仿真。我们展示了NIVA如何从贝叶斯推断视角统一现有的序列到序列轨迹预测模型与基于下一词预测（NTP）训练的新兴闭环仿真模型。在Waymo Open Motion Dataset上的实验表明，NIVA在性能上可与现有方法竞争，并提供对意图和驾驶风格的精细控制。
原文摘要 · Abstract (English)
The rapid iteration of autonomous vehicle (AV) deployments leads to increasing needs for building realistic and scalable multi-agent traffic simulators for efficient evaluation. Recent advances in this area focus on closed-loop simulators that enable generating diverse and interactive scenarios. This paper introduces Neural Interactive Agents (NIVA), a probabilistic framework for multi-agent simulation driven by a hierarchical Bayesian model that enables closed-loop, observation-conditioned simulation through autoregressive sampling from a latent, finite mixture of Gaussian distributions. We demonstrate how NIVA unifies preexisting sequence-to-sequence trajectory prediction models and emerging closed-loop simulation models trained on Next-token Prediction (NTP) from a Bayesian inference perspective. Experiments on the Waymo Open Motion Dataset demonstrate that NIVA attains competitive performance compared to the existing method while providing embellishing control over intentions and driving styles.
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。