TrajAir
收藏资源简介:
TrajAir数据集是由卡内基梅隆大学机器人学院创建,专注于非塔楼终端空域的航空轨迹预测。该数据集包含111天的ADS-B应答器数据和相应的METAR气象数据,旨在为自主航空提供首个3D社会空中导航数据集。数据集的创建过程涉及对ADS-B接收器收集的数据进行处理,以适应网络训练。TrajAir数据集的应用领域主要集中在解决非塔楼终端空域中飞机轨迹的预测问题,特别是在动态环境中的社会导航行为。
The TrajAir dataset was developed by the Robotics Institute at Carnegie Mellon University, focusing on aircraft trajectory prediction in non-towered terminal airspace. It includes 111 days of ADS-B transponder data and corresponding METAR meteorological data, aiming to serve as the first 3D social air navigation dataset for autonomous aviation. The dataset creation process involves processing data collected by ADS-B receivers to adapt to network training. The application scenarios of the TrajAir dataset mainly focus on addressing the aircraft trajectory prediction problem in non-towered terminal airspace, particularly for social navigation behaviors in dynamic environments.




