Abstract: To summarize the current situation of the fusion technology of visual sensors and LiDAR, this article focuses on the demand for more accurate obstacle recognition of autonomous vehicles during road driving. Firstly, the recognition principles and characteristics of visual sensors and light detection and LiDAR are introduced. By fusing the data layer, feature layer or decision-making layer of visual sensors and LiDAR, their respective deficiencies are compensated to improve the reliability and adaptability of the system. Secondly, the key technologies in the data fusion process are expounded, and the obstacle recognition algorithms and their improvement situations are sorted out. Finally, the challenges faced by this technology are summarized, aiming to provide comprehensive and in-depth technical references for researchers in related fields and promote further innovation and breakthroughs in this technology.

Key words: visual sensor; LiDAR; integration; obstacle recognition

摘要： 为总结视觉传感器和激光雷达（LiDAR）融合技术的现状，文章针对自动驾驶汽车在 道路行驶中需要更准确的障碍物识别需求。首先介绍了视觉传感器和激光雷达的识别原理及 特性，通过对视觉传感器和激光雷达融合数据层、特征层或决策层来弥补各自缺陷，提高系 统可靠性与适应性。其次阐述了数据融合过程的关键技术，梳理了障碍物识别算法及其改进 情况。最后对该技术面临的挑战进行总结，旨在为相关领域的研究人员提供全面且深入的技 术参考，推动该技术的进一步创新与突破。

关键词: 视觉传感器；激光雷达；融合；障碍物识别