Ding Zhigang, Gong Yangyang, Lin Jincheng, Jiang Jingjing, Zhang Jiadi
School of Mechanical and Automotive Engineering, Fujian University of Technology, Fuzhou, 350118, China.
Xiamen Richen Technology Co., Ltd., Xiamen, 361000, China.
Sci Rep. 2025 Jul 1;15(1):20806. doi: 10.1038/s41598-025-07269-y.
Sensor calibration is a crucial step to ensure the proper functioning of advanced driver assistance systems (ADAS), and vehicle coordinate system reconstruction is an indispensable part of this process. However, the currently mainstream vehicle alignment platforms are characterized by high costs, complex structures, and immobility, making them unsuitable for the rapid upgrade and transformation pace of modern automotive manufacturers. To address these limitations, this paper proposes a low-cost, flexible solution based on extracting key feature points from the car body-in-white. First, the paper introduces the reconstruction principles based on rigid body transformation. Then, the method for selecting the necessary feature points and its general applicability are discussed. Third, the paper describes how to mitigate the influence of vehicle body color by choosing appropriate light sources, and it employs suitable algorithms to complete the feature point extraction process. Finally, validation experiments were designed to verify the accuracy of the proposed method, demonstrating that it achieves better reconstruction precision compared to the vehicle alignment platforms.
传感器校准是确保高级驾驶辅助系统（ADAS）正常运行的关键步骤，而车辆坐标系重建是这一过程中不可或缺的一部分。然而，当前主流的车辆校准平台存在成本高、结构复杂和不可移动等特点，使其不适用于现代汽车制造商快速升级和转型的步伐。为了解决这些限制，本文提出了一种基于从白车身提取关键特征点的低成本、灵活的解决方案。首先，本文介绍了基于刚体变换的重建原理。然后，讨论了选择必要特征点的方法及其普遍适用性。第三，本文描述了如何通过选择合适的光源来减轻车身颜色的影响，并采用合适的算法完成特征点提取过程。最后，设计了验证实验来验证所提方法的准确性，结果表明与车辆校准平台相比，该方法具有更高的重建精度。
