基于机器人操作系统的自动驾驶小车系统研究  

Research on a ROS-based Autonomous Driving Vehicle System

作　　者：蒋红梅 黄鹏 唐军 JIANG Hongmei;HUANG Peng;TANG Jun

机构地区：[1]桂林信息科技学院,广西桂林541004 [2]桂林市自来水有限公司,广西桂林541004

出　　处：《科技创新与应用》2023年第17期46-49,共4页Technology Innovation and Application

基　　金：2022年度广西高校科学技术研究项目(2022KY1626)。

摘　　要：现阶段的导航电子地图的数据精度与实际行车过程需求的精度存在一定的差异,为了提供超高精度的导航地图与超高精度的物体识别,提出基于机器人操作系统(ROS)的自动驾驶小车系统设计。系统基于全球定位系统和激光雷达结合的粗/精准定位技术,采用ROS系统并结合SLAM算法及融合多种传感器,实现障碍物的高精度实时检测和精准定位建图。测试结果表明,基于ROS的自动驾驶小车具有实时路况及移动物体的高精度感知,进而能实现自动识别、自动导航、自动避障功能。基于ROS的自动驾驶小车系统能有效增加小车驾驶的舒适性和安全性,实现对小车驾驶控制的支持。Based on the fact that there is a certain difference between the data accuracy of the current navigation electronic map and the accuracy required by the actual driving process,a ROS-based autonomous driving vehicle system design is proposed in order to provide ultra-high-precision high-altitude maps and ultra-high-precision object recognition.Based on the rough/precise positioning technology combined with global positioning system and LiDAR,the system adopts ROS system,SLAM algorithm and fusion of various sensors to achieve high-precision real-time detection and precise positioning mapping of obstacles.The test results show that the autonomous vehicles based on ROS have the high-precision perception of real-time road conditions and moving objects and can realize the functions of automatic identification,automatic navigation and automatic obstacle avoidance.The ROS-based autonomous driving car system can effectively increase the comfort and safety of car driving and realize the support for car driving control.

关 键 词：自动驾驶 ROS 激光雷达 SLAM算法 GPS技术 

分 类 号：TP23[自动化与计算机技术—检测技术与自动化装置]