基于序列2D结构化线特征的桥梁检测机器人末端位姿估计方法  被引量:8

Estimation method of collecting terminal pose of bridge detection robot based on sequence 2D structured line features

在线阅读下载全文

作  者:刘理 王耀南 万智 何小娟 

机构地区:[1]湖南大学电气与信息工程学院机器人视觉感知与控制技术国家工程实验室 [2]湖南桥康智能科技有限公司

出  处:《电子测量与仪器学报》2018年第8期59-68,共10页Journal of Electronic Measurement and Instrumentation

基  金:国家自然科学基金(61733004,61573134,61433016,61401046);国家科技支撑计划(2015BAF11B01,2015BAF13B00);湖南省教育厅资助科研(17C0046);长沙市科技计划(K1404019-11)资助项目

摘  要:针对桥梁底部缺陷检测的人工作业难题,设计了智能移动式桥梁检测机器人系统,结合桥梁的结构化线特征和线轮廓模型2D结构特征,研究了桥梁检测机器人采集末端的位姿估计系统。首先,基于Hough变换和直线段检测(LSD)的结构化线特征的检测方法,提出了一种直线段合并、结构化线特征的角结构特征和结构化轮廓特征的检测算法。然后,在序列2D结构化线特征检测的基础上,进行了采集末端位姿估计算法的研究,利用假设-检验框架实现了初始位姿估计,提出了一种新的直线距离度量规则,设计了一种加权最小二乘算法实现采集末端连续位姿的实时估计。最后,开发出基于树莓派和Hokuyo单线激光器的硬件系统,并对提出的基于序列2D结构化线特征末端位姿估计方法进行了现场测试与验证,实验结果表明,该方法具有很好的鲁棒性,在实时性以及精度等方面均满足实际桥梁检测的需要。Aiming at the manual work problem of bridge bottom defect detection,an intelligent mobile bridge inspection robot system is designed. Combining with the structural line characteristics and 2 D structure characteristics of the linear contour model,the pose estimation system of the terminal of bridge detection robot is studied. First,based on the detection method of structured feature by Hough transform and LSD,a detection algorithm for the angle structure features and structured contour features is proposed by using line segment merging and structured line feature. Then,on the basis of the feature detection of the sequence 2 D structured line,a position estimation algorithm of the collection terminal is researched,an initial pose estimation is realized by using hypothesis testing framework,a new linear distance measurement rule is proposed,a real-time estimation of the weighted least squares algorithm to achieve continuous acquisition end pose is designed. Finally,the hardware system based on the raspberry and Hokuyo laser is developed and the proposed method is tested and verified in the field. The experimental results show that this method has good robustness and meets the needs of actual bridge detection in real time and accuracy.

关 键 词:桥梁检测机器人 结构化线特征 位姿估计 加权最小二乘算法 

分 类 号:TP391.41[自动化与计算机技术—计算机应用技术] TP242[自动化与计算机技术—计算机科学与技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象