TWICE Dataset: Digital Twin of Test Scenarios in a Controlled Environment

We have developed a comprehensive dataset that includes camera, radar, LiDAR, inertial measurement unit (IMU), and GPS sensor data recorded under adverse weather conditions in a controlled environment. This dataset also includes the digital twin of each test scenario made in a Hardware-in- the-Loop setup. We recorded test scenarios, some of which are catalogued by EURONCAP (European New Car Assessment Programme), using objects of interest such as car, cyclist, truck and pedestrian. All scenarios were recorded in the outdoor facilities of the CARISSMA Institute of Automated Driving in Ingolstadt, Germany. The dataset contains more than 2 hours of recording, which totals more than 280GB of data. Overall, this dataset provides a valuable resource for researchers in the field of autonomous vehicles to test and improve their algorithms in adverse weather conditions, as well as explore the simulation-to-reality gap.