首页|基于知识图谱的无人驾驶车辆目标检测研究现状及热点分析

基于知识图谱的无人驾驶车辆目标检测研究现状及热点分析

扫码查看
随无人驾驶汽车智能等级不断提升,对目标检测精度的要求不断提高,有必要系统全面地梳理无人驾驶车辆目标检测领域的研究现状和前沿方向.本文利用知识图谱可视化软件Citespace在CNKI数据库中以"目标检测"&"车辆"为篇名进行文献检索分析,发现无人驾驶车辆目标检测研究热点主要聚焦于:基于人工智能辅助的智能车辆技术,基于深度学习算法的多元数据特征融合技术,基于点云数据目标检测图像深度补全技术,及基于深度学习算法的小目标检测技术.在分析研究热点的基础上通过"突现词"发现提升深度学习算法在复杂环境下低分辨率目标的适应性,及基于深度学习的自动驾驶技术数据轻量化、运算快速化和目标检索准确化是当前研究前沿问题.
Research Status and Hotspots Analysis for Object Detection of Driverless Vehicles Based on Knowledge Graph
With the continuous improvement of the intelligent level of driverless vehicles and the increasing demand for object detection accuracy,it is necessary to systematically and comprehensively review the research status and frontier directions in the field of object detection for driverless vehicles.This paper utilized the knowledge graph visualization software Citespace to conduct a literature search and analysis in the CNKI database using the keywords"object detection"and"vehicle".The result showes that the frontier directions of object detection for driverless vehicles mainly focus on:intelligent vehicle technology based on artificial intelligence assistance,multi-data feature fusion technology based on deep learning algorithms,image depth completion technology for object detection based on point cloud data,and small object detection technology based on deep learning algorithms.Based on the analysis of research hotspots,it is discovered through"burst terms"that enhancing the adaptability of deep learning algorithms to low-resolution objects in complex environments,as well as lightweighting data,accelerating computation,and improving target retrieval accuracy in autonomous driving technology based on deep learning,are currently the forefront issues in research.

driverless vehiclesobject detectionknowledge graphmultiple sensor fusiondeep learning algorithm

魏海姣、郑凯锋、李璞、刘伟、支博文、曹柯

展开 >

中国北方车辆研究所,北京100072

内蒙古第一机械集团股份有限公司,包头014032

无人驾驶 目标检测 知识图谱 多传感器融合 深度学习

2024

车辆与动力技术
中国兵工学会

车辆与动力技术

影响因子:0.287
ISSN:1009-4687
年,卷(期):2024.(3)