杨静静 曹海平 丁杰 顾银波 霍旋
（南通大学张謇学院，江苏 南通 226000）
摘要：针对传统的自主循迹避障小车在复杂路况下存在图像识别精度低、处理时间长等问题，设计一种基于EdgeBoard的智能循迹避障小车。首先，通过USB摄像头采集道路图像；然后，采用基于OpenCV的道路循迹算法识别道路边缘；最后，利用SSD_MobileNetV1深度学习模型识别特定的交通标志。经测试，该智能循迹避障小车具有良好的稳定性和可靠性。
关键词：循迹避障小车；EdgeBoard；OpenCV；深度学习；自主循迹避障
中图分类号：TP242.6 文献标志码：A 文章编号：1674-2605(2024)02-0004-07
DOI：10.3969/j.issn.1674-2605.2024.02.004
Design of Intelligent Tracking and Obstacle Avoidance Vehicle Based on EdgeBoard
YANG Jingjing CAO Haiping DING Jie GU Yinbo HUO Xuan
(Zhangjian School, Nantong University, Nantong 226000, China)
Abstract: Aiming at the problems of low image recognition accuracy and long processing time in traditional autonomous obstacle avoidance vehicles under complex road conditions, a smart obstacle avoidance vehicle based on EdgeBoard is designed. Firstly, collect road images through a USB camera; Then, the road tracking algorithm based on OpenCV is used to identify the edges of the road; Finally, use the SSD_MobileNetV1 deep learning model to identify specific traffic signs. After testing, the intelligent tracking and obstacle avoidance car has good stability and reliability.
Keywords: tracking and obstacle avoidance vehicle; EdgeBoard; OpenCV; deep learning; autonomous tracking and obstacle avoidance
