TWICE Dataset
收藏资源简介:
TWICE数据集是由CARISSMA自动驾驶研究所开发的,旨在为自动驾驶车辆在恶劣天气条件下的安全可靠操作提供支持。该数据集包含在真实测试赛道上收集的传感器数据,并在实验室中重现相同的测试场景。数据集包括相机、雷达、激光雷达、惯性测量单元和GPS数据,记录了在雨天、夜间和雪天等恶劣天气条件下的情况。数据集总记录时长超过2小时,数据量超过280GB,适用于自动驾驶领域的研究人员测试和改进算法,以及探索模拟与现实之间的差距。
The TWICE dataset was developed by the CARISSMA Institute of Autonomous Driving to support the safe and reliable operation of autonomous vehicles under adverse weather conditions. The dataset encompasses sensor data collected on real-world test tracks, alongside replicated versions of the same test scenarios conducted in a laboratory environment. It includes data from cameras, radar, LiDAR, inertial measurement units (IMUs), and GPS devices, recording scenarios under adverse weather conditions such as rainy days, nighttime, and snowy days. The total recording duration of the dataset exceeds 2 hours, with an overall data volume surpassing 280 GB. It is suitable for researchers in the autonomous driving domain to test and optimize their algorithms, as well as to investigate the gap between simulation and real-world scenarios.

- 1TWICE Dataset: Digital Twin of Test Scenarios in a Controlled EnvironmentCARISSMA自动驾驶研究所 (C-IAD) · 2023年



