摘要
当前,变电巡检装置以视觉机器人为主,易受到外界环境的影响,导致自主避障控制效果不理想,对此,提出一种基于机器视觉的变电巡检装置自主避障控制方法。利用最小错误率贝叶斯决策对图像分割处理,通过Hough变换进行道路检测,提取无障碍物路面区域,同时展开路面分割,根据分割结果确定障碍物曲面坐标。设计基于小波神经的网络滑膜控制器,通过控制器指导机器人躲避障碍物,实现变电巡检装置自主避障控制方法。实验结果表明,这种方法可以为变电巡检装置提供有效的避障信息,自主避障控制时间在0.07 s以下,具有良好的自主避障控制效果。
At present,the substation patrol inspection device is mainly a visual robot,which is vulnerable to the influence of the external environment,resulting in the unsatisfactory effect of autonomous obstacle avoidance control.An autonomous obstacle avoidance control method based on machine vision for substation patrol inspection device is proposed.The minimum error rate Bayesian decision is used to process the image segmentation,and the Hough transform is used to detect the road,extract the barrier-free pavement area,carry out the pavement segmentation at the same time,and determine the obstacle surface coordi-nates according to the segmentation results.A sliding film controller based on wavelet neural network is designed to guide the robot to avoid obstacles and realize the autonomous obstacle avoidance control method of the substation patrol inspection device.The experimental results show that this method can provide effective obstacle avoidance information for the substation patrol in-spection device.The autonomous obstacle avoidance control time is less than 0.07 s,and has good autonomous obstacle avoid-ancecontrol effect.
作者
黄炜昭
陈龙
何维
谢欢欢
辛拓
张宏钊
HUANG Weizhao;CHEN Long;HE Wei;XIE Huanhuan;XIN Tuo;ZHANG Hongzhao(Shenzhen Power Supply Bureau Co.,Ltd.,Shenzhen 518000,China)
出处
《微型电脑应用》
2025年第5期83-87,共5页
Microcomputer Applications
基金
深圳供电局有限公司(090000KK52200150)。
关键词
机器视觉
变电巡检装置
自主避障控制
图像分割
machine vision
substation patrol inspection device
autonomous obstacle avoidance control
image segmentation
