机器人抓取视觉传感目标精确定位方法  被引量:2

Precise Positioning Method of Robot Grasping Visual Sensing Target

在线阅读下载全文

作  者:沈微微[1,3] 姜晓燕 张兵 郭新年[1] SHEN Weiwei;JIANG Xiaoyan;ZHANG Bing;GUO Xinnian(Institute of Information Engineering,Suqian University,Suqian Jiangsu 223800,China;School of Electronic and Electrical Engineering,Shanghai University of Engineering Science,Shanghai 201620,China;Jiangsu Province Engineering Research Center of Smart Poultry Farming and Intelligent Equipment,Suqian Jiangsu 223800.China)

机构地区:[1]宿迁学院信息工程学院,江苏宿迁223800 [2]上海工程技术大学电子电气工程学院,上海201620 [3]江苏省家禽智慧养殖与智能装备工程研究中心,江苏宿迁223800

出  处:《传感技术学报》2023年第5期763-768,共6页Chinese Journal of Sensors and Actuators

基  金:国家自然科学基金委青年资助项目(62001183,61702322);教育部产学合作协同育人项目(202102645020);江苏省教育科学“十四五”规划课题项目(B/2021/01/80);宿迁市科技计划项目(L202109);国家自然科学基金,民航联合基金重点项目(U2033218)。

摘  要:机器人抓取目标时,正确完成任务的前提是可以精准检测到目标位置,当距离目标较远时,以信号传感为基础的定位精度和稳定性会受到影响。为解决上述问题,提出基于视觉传感器的机器人抓取目标精确定位方法。利用视觉传感器获取目标图像,并标定目标位姿。采取直线段检测方法提取目标位姿特征,将提取的特征输入到改进粒子群算法的支持向量机回归模型中,输出定位结果。利用回归误差补偿模型对定位结果补偿,完成机器人抓取目标精确定位。实验结果显示,利用视觉传感器后,机器人抓取目标的定位时间为35 s、与实际位置的接近程度高于81%、置信度高于92%,由此可知机器人抓取视觉传感目标定位效果较好。When the robot grabs the target,the premise of correctly completing the task is that it can accurately detect the target position.When it is far from the target,the positioning accuracy and stability based on signal sensing will be affected.In order to solve the above problem,an accurate positioning method of robot grasping target based on vision sensor is proposed.Vision sensor is used to obtain the target image and calibrate the target pose.The linear segment detection method is used to extract the target pose features,the extracted features are input into the support vector machine regression model of the improved particle swarm optimization algorithm,and the positioning results are output.The regression error compensation model is used to compensate the positioning results to complete the accurate positioning of the robot grasping target.The experimental results show that after using the visual sensor,the positioning time of the robot grasping the target is 35 s,the proximity to the actual position is higher than 81%,and the confidence is higher than 92%.Therefore,it can be seen that the visual sensor target positioning effect of robot grasping is better.

关 键 词:视觉传感器 直线段检测算法 标定目标位姿 粒子群算法 误差补偿模型 惩罚因子 

分 类 号:TN911.73[电子电信—通信与信息系统]

 

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