📄 中文摘要
监控路线和轨道环境在自动驾驶中扮演着重要角色，尤其是在自动化等级为2的情况下，列车司机仍在车上。随着自动化等级达到4，完全无人驾驶时，这些系统将完全独立地进行环境监控。利用人工智能（AI），系统能够自动对路线上的风险和危险事件做出反应。为了训练这些AI算法，需要大量的高质量训练数据，这些数据因其安全性的重要性而必须达到高标准。研究提出了一种自动化的方法，用于确保训练数据的质量，旨在提高自动驾驶系统的安全性和可靠性。
传感器数据注释的自动质量检查
出处: Automated Quality Check of Sensor Data Annotations
发布: 2026年3月3日
📄 English Summary
Automated Quality Check of Sensor Data Annotations
Monitoring the route and track environment plays a crucial role in automated driving, particularly at Grade of Automation (GoA) level 2, where the train driver is still present. At fully automated, driverless operation at GoA level 4, these systems take over environment monitoring completely independently. With the aid of artificial intelligence (AI), they can automatically respond to risks and hazardous events on the route. To train such AI algorithms, large volumes of high-quality training data are required, which must meet stringent safety standards. This study presents an automated method for ensuring the quality of training data, aimed at enhancing the safety and reliability of automated driving systems.