CAJ | 학술논문

研究了基于智能算法的机器人无标定视觉伺服问题,提出了一种新的基于最小二乘支持向量回归的机器人无标定视觉免疫控制方法.利用最小二乘支持向量回归学习机器人位姿变化和观测到的图像特征变化之间的复杂非线性关系,其中最小二乘支持向量回归的参数由自适应免疫算法加5折交叉检验优化确定,在此基础上利用免疫控制原理设计了视觉控制器.六自由度工业机器人空间4DOF视觉定位实验结果表明了该方法的有效性.
연구료기우지능산법적궤기인무표정시각사복문제,제출료일충신적기우최소이승지지향량회귀적궤기인무표정시각면역공제방법.이용최소이승지지향량회귀학습궤기인위자변화화관측도적도상특정변화지간적복잡비선성관계,기중최소이승지지향량회귀적삼수유자괄응면역산법가5절교차검험우화학정,재차기출상이용면역공제원리설계료시각공제기.륙자유도공업궤기인공간4DOF시각정위실험결과표명료해방법적유효성.
The robot uncalibrated visual servoing based on an intelligent algorithm is studied; and a new robot uncal-ibrated visual servoing method based on the least squares support vector regression(LS-SVR) is proposed. LS-SVR is used to learn the complex nonlinear relationship between the changes of robot pose and the variations of the image features observed. The parameters of LS-SVR are determined by the adaptive immune algorithm(AIA) plus a 5-fold cross valida-tion: and the visual controller is designed using the principle of immune feedback. 4DOF visual positioning experiments have been carded out in a 6DOF industrial robot. Experimental results testify the feasibility and validity of this method.