F1TENTH: An Open-source Evaluation Environment for Continuous Control and Reinforcement Learning
F1TENTH: An Open-source Evaluation Environment for Continuous Control and Reinforcement Learning
复制标题
DOI：
--
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Matthew O'Kelly;Hongrui Zheng;D. Karthik;Rahul Mangharam
中科院分区：
文献类型：
--
作者：
Matthew O'Kelly;Hongrui Zheng;D. Karthik;Rahul Mangharam
在自动驾驶汽车（AV）上部署和评估学习算法是昂贵、缓慢且潜在不安全的。本文详细介绍了f110自主赛车平台，这是一个用于训练、测试和评估自主系统的开源评估框架。凭借1/10的低成本硬件和多个虚拟环境，f110th即使在实验室研究环境中也能安全快速地进行自动驾驶算法实验。我们在自动驾驶赛车的背景下提出了三个基准任务和基线，展示了我们的评估环境的灵活性和特征
The deployment and evaluation of learning algorithms on autonomous vehicles (AV) is expensive, slow, and potentially unsafe. This paper details the F1TENTH autonomous racing platform, an open-source evaluation framework for training, testing, and evaluating autonomous systems. With 1/10th-scale low-cost hardware and multiple virtual environments, F1TENTH enables safe and rapid experimentation of AV algorithms even in laboratory research settings. We present three benchmark tasks and baselines in the setting of autonomous racing, demonstrating the (cid:13)exibility and features of our evaluation environment