A collaborative team from the School of Mechanical and Vehicular Engineering and the National Key Laboratory of Intelligent Unmanned Systems Technology at Beijing Institute of Technology (BIT), alongside researchers from Tsinghua University and the National University of Singapore, has unveiled a groundbreaking Dual Radar dataset for autonomous driving in complex environments. This pioneering work, published in a Nature sub-journal, utilizes 4D imaging millimeter-wave technology. The dataset boasts over 400 kilometers of perception data captured under non-ideal conditions, encompassing more than 10,000 synchronized frames with annotated 3D object detection and tracking data. Its primary objective is to rigorously evaluate the efficacy of current perception algorithms in real-world, less-than-ideal scenarios.
