comma2k19
收藏资源简介:
comma2k19数据集由comma.ai创建,包含2019个1分钟长的视频片段,总计超过33小时的高速公路驾驶数据。数据集通过comma EON设备收集,包含多种传感器数据,如摄像头、GPS、温度计和IMU等。这些数据集中在一个20公里的加州高速公路上,旨在通过使用普通传感器来开发和验证紧密耦合的GNSS算法和地图算法。数据集特别适用于开发高性能的定位和地图算法,这些算法旨在在普通硬件上运行,解决自动驾驶等领域的问题。
The comma2k19 dataset, created by comma.ai, comprises 2019 one-minute video clips, totaling over 33 hours of highway driving footage. Collected via comma EON devices, the dataset encompasses diverse sensor data including cameras, GPS receivers, thermometers, and inertial measurement units (IMUs), among others. All data was gathered along a 20-kilometer stretch of highway in California. Its core objective is to develop and validate tightly coupled GNSS algorithms and mapping algorithms using commodity off-the-shelf sensors. This dataset is particularly well-suited for developing high-performance positioning and mapping algorithms designed to run on standard hardware to address challenges in domains such as autonomous driving.




