The Oxford RobotCar Dataset contains over 100 repetitions of a consistent route through Oxford, UK, captured over a period of over a year. The dataset captures different combinations of weather, traffic and pedestrians, along with longer term changes such as construction and roadworks. The dataset includes raw recordings from all the sensors, a full set of intrinsic and extrinsic sensor calibrations, and MATLAB development tools for handling the raw sensor data.
The dataset is intended for research on long-term autonomous driving, specifically localization and mapping for autonomous vehicles in real-world, dynamic urban environments.
Verified
@article{RobotCarDatasetIJRR,
author = {Will Maddern and Geoff Pascoe and Chris Linegar and Paul Newman},
doi = {10.1177/0278364916679498},
eprint = {http://ijr.sagepub.com/content/early/2016/11/28/0278364916679498.full.pdf+html},
journal = {The International Journal of Robotics Research (IJRR)},
number = {1},
pages = {3-15},
pdf = {http://robotcar-dataset.robots.ox.ac.uk/images/robotcar_ijrr.pdf},
title = {{1 Year, 1000km: The Oxford RobotCar Dataset}},
url = {http://dx.doi.org/10.1177/0278364916679498},
volume = {36},
year = {2017}
}
