期刊文献+

改进PROSAC算法的并联机器人末端位姿检测方法 被引量:1

POSE DETECTION FOR PARALLEL ROBOT BASED ON IMPROVED PROSAC ALGORITHM
下载PDF
导出
摘要 为解决双目视觉末端位姿检测中光照、噪声干扰等外部因素造成的检测精度降低问题,提出一种改进PROSAC (Progressive Sample Consensus)算法的水果分拣并联机器人双目视觉末端位姿检测方法。基于ORB算法进行特征提取和立体匹配;采用改进的PROSAC算法对立体匹配进行提纯,该改进通过穿插取点和预检验候选模型克服PROSAC算法存在的模型参数估计精度不高和验证错误候选模型耗时问题;将提纯后的匹配点对代入双目视觉模型求出末端位姿。实验结果表明,与未改进PROSAC算法的末端位姿检测方法相比,改进PROSAC算法的并联机器人末端位姿检测方法,其位姿各分量x、y、z、γ的误差平均绝对值分别降低了53.9%、65.5%、66.9%、47%,误差标准差分别降低了53.2%、67%、66.6%、56.6%,验证了所提出方法的有效性。 In the pose detection for a fruit sorting parallel robot based on binocular vision, the precision of pose detection is adversely affected by external factors such as illumination and noise interference. To solve the above problems, a pose detection method for the fruit sorting parallel robot based on the improved PROSAC algorithm is proposed. ORB algorithm was adopted to extract feature points and match images. An improved PROSAC algorithm was proposed to purify the results of matching, which could overcome the low estimation accuracy with respect to the model parameters and time-consuming on testing the error candidate model via selecting feature points separately and pre-testing candidate model. The purified matching points were taken into the binocular vision model to calculate the pose parameter. Experimental results show that compared with the pose detection method based on the unimproved PROSAC algorithm, when the pose detection method based on the improved PROSAC algorithm is applied, the mean absolute value of error for pose component x, pose component y, pose component z and pose component γ are reduced by 53.9%, 65.5%, 66.9% and 47% respectively;the standard deviation of error for pose component x, pose component y, pose component z and pose component γ are reduced by 53.2%, 67%, 66.6% and 56.6% respectively. The results verify the effectiveness of the proposed method.
作者 高国琴 韩滢 Gao Guoqin;Han Ying(School of Electrical and Information Engineering,Jiangsu University,Zhenjiang 212013,Jiangsu,China)
出处 《计算机应用与软件》 北大核心 2022年第4期205-212,262,共9页 Computer Applications and Software
基金 国家自然科学基金项目(51375210) 镇江市重点研发计划项目(GZ2018004) 江苏高校优势学科建设工程资助项目。
关键词 水果分拣并联机器人 双目视觉 PROSAC 位姿检测 A fruit sorting parallel robot Binocular vision PROSAC(Progressive Sample Consensus) Pose detection
  • 相关文献

参考文献10

二级参考文献78

共引文献132

同被引文献16

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部