室外广域环境下移动机器人地形障碍检测及分析方法  

Terrain Obstacle Detection and Analysis Method for Mobile Robot in Outdoor Wide-area Environment

作　　者：韩广涛 张明路[1] 高春艳[1] 吕晓玲[1] HAN Guang-tao;ZHANG Ming-lu;GAO Chun-yan;LÜXiao-ling(School of Mechanical Engineering,Hebei University of Technology,Tianjin 300401,China)

机构地区：[1]河北工业大学机械工程学院,天津300401

出　　处：《科学技术与工程》2024年第19期8150-8157,共8页Science Technology and Engineering

基　　金：国家重点研发计划(2022YFB4701104);国家自然科学基金重点项目(U1913211)。

摘　　要：近年来,移动机器人已广泛应用于应急救援、公共安防、巡检探测等多个领域。在机器人移动过程中,环境感知能力尤为重要,尤其在包含动态、未知以及复杂因素干扰的室外广域工作场景中,障碍物检测极大影响机器人的功能及效率。针对地面移动机器人在此类场景下对障碍物检测的研究进展进行了对比,并根据障碍物的特征对正、负障碍检测方法进行了分析;进而概述了信息融合和深度学习在当前领域的应用现状;最后,总结了障碍物检测领域的研究现状,并指出相关技术的探索方向。In recent years,mobile robots have been widely used in various fields such as emergency rescue,public security,inspection and detection.In the process of robot movement,the ability to perceive the environment is particularly important,especially in outdoor wide area work scenes involving dynamic,unknown,and complex factors.Obstacle detection greatly affects the functionality and efficiency of the robot.A comparison was made on the research progress of obstacle detection for ground mobile robots in such scenarios,and positive and negative obstacle detection methods were analyzed based on the characteristics of obstacles.Furthermore,the current application status of information fusion and deep learning in the current field was summarized.Finally,the current research status in the field of obstacle detection was summarized,and the exploration direction of related technologies was pointed out.

关 键 词：移动机器人 非结构化场景 障碍物检测 信息融合 深度学习 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置] TP212[自动化与计算机技术—控制科学与工程]