基于双高斯过程的协作机器人自适应策略

发布时间：2018-04-15 10:42

本文选题：高斯过程 + 自适应策略　；参考：《计算机集成制造系统》2017年09期

【摘要】：为适应目标物体位姿的变化,实现运动的平滑性,提出一种机器人自适应策略,在位姿适应和轨迹调整两个环节上各建立一个高斯过程模型。位姿适应的高斯过程通过线性协方差函数将观测变量和机器人关节变量关联,避免了视觉系统的校正和机器人运动学逆解;轨迹调整的高斯过程利用高斯核函数计算关节轨迹点之间的协方差,使调整后的机器人轨迹更加平滑。通过UR3机器人在有障碍物下的自适应抓取实验,证明了所提方法既能够适应目标物体的位姿变化,又能得到平滑的关节运动。
[Abstract]:In order to adapt to the change of the position and pose of the target object and realize the smoothness of motion, a robot adaptive strategy is proposed, in which a Gao Si process model is established on the two links of pose adaptation and trajectory adjustment.Gao Si process of position and pose adaptation correlates observation variable with robot joint variable through linear covariance function, which avoids the correction of vision system and kinematics inverse solution of robot.The Gao Si process of trajectory adjustment calculates the covariance between the joint locus points by Gao Si kernel function, which makes the adjusted robot trajectory smoother.The adaptive grab experiment of UR3 robot with obstacles proves that the proposed method can not only adapt to the position and pose changes of the target object, but also obtain smooth joint motion.
【作者单位】：北京航空航天大学机械工程及自动化学院;
【基金】：国家科技支撑计划资助项目(2015BAF01B04) 北京市科技计划资助项目(D161100003116002)~~
【分类号】：TP242

【相似文献】