圆管作业机器人视觉惯导组合姿态定位研究  

Research on vision-inertial navigation of an underwater cleaning robot

作　　者：江平[1] 杨灿军[1] 寿志成 陈燕虎[1] 范锦昌[1] 黄政明 魏谦笑 JIANG Ping;YANG Canjun;SHOU Zhicheng;CHEN Yanhu;FAN Jinchang;HUANG Zhengming;WEI Qianxiao(State Key Laboratory of Fluid Power and Mechatronic Systems, Zhejiang University, Hangzhou 310027, China;Shanghai Branch of China National Offshore Oil Corporation, Shanghai 200335, China)

机构地区：[1]浙江大学流体动力与机电系统国家重点实验室,浙江杭州310027 [2]中海油能源发展装备技术有限公司上海分公司,上海200335

出　　处：《中南大学学报（自然科学版）》2018年第12期2946-2952,共7页Journal of Central South University:Science and Technology

基　　金：浙江省公益技术应用研究计划项目(2016C33057);中国海洋石油总公司科技项目(CNOOC-KJ125 ZDXM 13 LTD NFZB 2015-04)~~

摘　　要：针对石油平台导管架水下清洗机器人姿态定位问题,提出一种新型的视觉惯导组合导航系统。该系统利用Canny算子与Hough变换识别圆管特征直线,并通过建立机器人空间模型,解算机器人空间位置姿态角。针对视觉惯导系统模型,设计基于四元数的扩展卡尔曼滤波,实现视觉与惯导信息融合,减小空间角度的累计误差,提高组合导航精度。研究结果表明:本文提出的机器人组合导航系统角度精度高,其精确度及实时性满足了机器人实际应用需求。For the localization of the underwater cleaning robot’s localization,an innovative vision-inertial navigation system was put forward.By recognizing pipe line in Canny/Hough and establishing robot’s spatial model,the attitude angles were solved.For vision-inertial system model,the extended Kalman filter bases on quaternion was designed to fuse vision/inertial data and reduce accumulative errors,which improves the system accuracy.The results show that this attitude accuracy is high,which indicates that the real-time performance and the precision of the navigation system can meet the practical demand.

关 键 词：姿态定位 视觉导航 组合导航 图像识别 扩展卡尔曼滤波 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]