RobotCar Long-Term Autonomy Benchmark

The initial release of the RobotCar Dataset consisted purely of raw data, giving researchers a large collection of data with a wide range of variation in illumination, weather, dynamic objects, seasonal changes, roadworks and building construction.

Building on that release, today at the ICRA Workshop on Robotics and Vehicular Technologies for Self-driving cars, we announced that we will soon be releasing a localisation benchmarking service, based on RTK-corrected GPS/INS trajectories.

We plan to offer a subset of the corrected RTK solutions for a small number of traversals, and withhold the remaining ground truth in order to provide an online benchmarking service.

By providing a benchmark where researchers can quantitatively evaluate and compare localisation and mapping approaches on a challenging large-scale dataset, we hope to accelerate development of long-term autonomy for future autonomous vehicles.

Please enter your email address below if you wish to be notified when the benchmark becomes available.