This dataset will allow researchers across the world to develop safe autonomous driving technology, says Aptiv.
Aptiv Autonomous Mobility, one of the leading suppliers developing automated driving (AD) systems, has released a comprehensive open-source automated vehicle datset, dubbed as nuTonomy scenes (nuScenes).
The nuScenes dataset is a large-scale autonomous driving dataset that allows researchers to study challenging urban driving situations using the full sensor suite of a real self-driving car.
The company claims that nuScenes is the first large-scale dataset to provide data from the entire sensor suite of an autonomous vehicle (six cameras, five radars and one lidar, all with full 360 degree field of view).
“The nuScenes dataset is inspired by the pioneering KITTI dataset. Compared to KITTI, nuScenes includes 7x more object annotations,” it said.
Data collected from Boston and Singapore
According to the company, the nuScenes data has been collected from Singapore and Boston, the two cities that are known for their dense traffic and highly challenging driving situations.
The company has released the full nuScenes dataset with all 1000 driving scenes collected from Boston and Singapore. The full dataset includes approximately 1.4M camera images, 390,000 lidar sweep, 1.4M radar sweeps and 1.4M object bounding boxes in 40k keyframes.
It also plans to include more features like map layers, raw sensor data, etc. soon.
The nuScenes dataset is available as free to use strictly for non-commercial purposes. Aptiv is hoping that this dataset will allow researchers across the world to develop safe autonomous driving technology.
