发表机构
TU Wien; AIT Austrian Institute of Technology（维也纳技术大学; 奥地利技术研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究针对RoboRacer平台赛道扩展问题，开发RoboRacer Arena系统可从占用地图自动生成高保真3D赛车环境，支持自然语言生成赛道，在Isaac Sim中实现高效可复现的自主赛车仿真。
AI 中文摘要
RoboRacer为使用1:10比例自主车辆的研究提供了标准化平台，但可用赛道的多样性阻碍了策略获取过程。尽管现有的占用网格模拟器允许快速添加新地图，但它们未包含物理接触；而3D模拟器要求每个赛道作为单独资产实现，因此限制了可扩展性。为解决该问题，我们开发了RoboRacer Arena，这是一种可直接从占用地图创建3D赛车环境的系统。我们的方法首先使用洪水填充算法提取可驾驶通道并识别赛道边界，随后用于建立障碍物；计算距离场以定义碰撞边界；将赛道表面、碰撞属性和材料组装成USD stage，支持在Isaac Sim中自动且可复现地生成环境。输入地图可从SLAM会话、重新缩放的一级方程式（Formula 1）赛道或自然语言描述获取。当输入基于自然语言时，我们使用Gemma 4 31B生成赛道规范，无需指定任何坐标或几何信息。为保证一致性和可复现性，我们应用几何筛选、程序生成和光栅级验证。模拟环境的初始化时间范围为1.18至2.48秒，且初始化时间随光栅尺寸增加呈线性增长。在涉及10条赛道、3个随机种子的30次匹配试验中，共生成21张地图，全部通过验证。RoboRacer Arena当前包含130条赛道，支持从自然语言生成赛道。在基准测试中，使用256个并行刚体车辆时，系统达到每秒8707个车辆步，不包含渲染和策略执行所需时间。
英文摘要
Learning-based autonomous racing relies on diverse training environments, yet constructing a new track requires geometric design, validation, and simulation-asset generation. This coupling makes track geometry difficult to vary systematically during policy training and evaluation. To address this limitation, we present RoboRacer Arena, a specification-driven pipeline that exposes track geometry as an explicit experimental variable. Starting from natural-language requirements, a seeded coverage-guided constructor generates closed-loop layouts, validates the exported occupancy maps, and automatically builds the corresponding Isaac Sim environments. The same interface admits recorded maps and scaled circuits, yielding an initial reference collection of 130 tracks. Across our evaluation, RoboRacer Arena achieves the highest valid-map generation rate among the tested construction procedures under their respective computational budgets and converts eight benchmark maps into simulation assets in less than 2.5 seconds each. We further reconstruct the RoboRacer vehicle as a CAD and USD asset and use it for parallel residual-policy training and deployment on the physical platform. In physical experiments, policies trained with RoboRacer Arena complete ten consecutive laps at command settings up to four times the nominal training speed. These results demonstrate that explicit track requirements can be connected to validated simulation assets and physical evaluation within a reproducible autonomous-racing workflow.
CommentsSubmitted to ICRA 2027
