仿人多指手逆运动学及协同运动控制的研究

发布时间：2018-04-24 20:25

本文选题：多指手 + 逆运动学　；参考：《哈尔滨工业大学》2017年硕士论文

【摘要】：本文结合国家自然科学基金项目“仿人多指手的多维指尖力感知和同步控制”(课题编号51175106),对仿人多指手的逆运动学和协同运动控制展开了研究。多指手系统的整体性能取决于所有手指以及手指所有关节的协同性能,如果多指手协同性差,那么系统的整体性能就会降低。为解决上述问题,本文在提出具有耦合关节的手指逆运动学算法后,研究单手指各关节间的同步误差和指间同步控制方法,并把人手协同运动规律引入控制器的设计中,使得多指手达到协同运动的目的。本文首先针对具有耦合关节的手指,提出基于幂级数展开的逆运动学算法,将逆运动学中遇到的带有三角超越函数的非线性问题转化为代数问题。并采用一维线性插值补偿误差,降低了对非线性耦合关节传动比选择准确性的要求。然后以HIT/DLR Hand II灵巧手为实验平台,分析不同空间轨迹规划对位置控制的影响,研究单手指各关节间的同步误差,引入基于卡尔曼滤波器的自适应摩擦力补偿方法,实现自适应摩擦力补偿。最后提出基于人手协同运动规律的多指手同步控制方法,在规划层引入基于人手模块划分的关节耦合比,在控制层设计不仅考虑到各手指关节和指尖位置的同步性,还包含人手协同运动规律的指间同步误差,并通过阻抗控制实验表明此方法能够较好复现人手典型抓取动作,实现协同运动。
[Abstract]:In this paper, the inverse kinematics and cooperative motion control of humanoid multi-fingered hand are studied in combination with the project of National Natural Science Foundation of China "perception and synchronous control of multi-finger-like force of humanoid multi-fingered hand" (subject number 51175106). The overall performance of the multi-fingered hand system depends on the cooperative performance of all fingers and all joints of the fingers. If the multi-fingered hand cooperation is poor, the overall performance of the system will be reduced. In order to solve the above problems, this paper presents an inverse kinematics algorithm of fingers with coupling joints, and studies the synchronization error and synchronization control method between the joints of one hand and fingers, and introduces the law of hand coordination motion into the design of the controller. Make the multi-fingered hand achieve the purpose of cooperative movement. In this paper, an inverse kinematics algorithm based on power series expansion is proposed for fingers with coupling joints. The nonlinear problem with trigonometric transcendental function encountered in inverse kinematics is transformed into an algebraic problem. One dimensional linear interpolation is used to compensate the error, which reduces the accuracy of the selection of the nonlinear coupling joint transmission ratio. Then, on the platform of HIT/DLR Hand II dexterous hand, the influence of different space trajectory planning on position control is analyzed, and the synchronization error between the joints of one hand finger is studied. An adaptive friction compensation method based on Kalman filter is introduced. Adaptive friction compensation is realized. Finally, a multi-fingered hand synchronization control method based on the law of human hand cooperative motion is proposed. The joint coupling ratio based on manual module division is introduced in the planning layer. In the control layer, not only the synchronization of each finger joint and finger tip is considered in the design of the control layer. It also includes the inter-finger synchronization error of the law of human hand cooperative motion, and the impedance control experiment shows that this method can reproduce the typical grasp action of the hand better and realize the cooperative motion.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242

【参考文献】