Learning Global Direct Inverse Kinematics
Learning Global Direct Inverse Kinematics
复制标题
DOI：
--
复制
发表时间：
1991-12
期刊：
影响因子：
56.9
通讯作者：
D. DeMers;K. Kreutz-Delgado
中科院分区：
文献类型：
--
作者：
D. DeMers;K. Kreutz-Delgado
我们介绍并证明了一个引导方法的机器人运动学映射的反函数的建设只使用样本配置-空间/工作空间数据。无监督学习（聚类）技术用于前图像邻域，以学习将配置空间划分为运动映射可逆的子集。然后在每个分区上单独使用监督学习来近似逆函数。由此正则化了不适定的逆运动学函数，并给出了无腕Puma机器人的全局逆运动学解。
We introduce and demonstrate a bootstrap method for construction of an inverse function for the robot kinematic mapping using only sample configuration--space/ workspace data. Unsupervised learning (clustering) techniques are used on pre-image neighborhoods in order to learn to partition the configuration space into subsets over which the kinematic mapping is invertible. Supervised learning is then used separately on each of the partitions to approximate the inverse function. The ill-posed inverse kinematics function is thereby regularized, and a global inverse kinematics solution for the wristless Puma manipulator is developed.