【作者】 刘兵；

【导师】 王刚；

【作者基本信息】 大连理工大学 ， 工程力学， 2023， 硕士

【摘要】 仿人多指手作为智能设备末端操作机构,具有媲美人手的外观结构和运动灵活性,能够完成对众多物体的抓取操作,为人类的生产活动和日常生活提供了强大的技术支撑和极大的便利。为了追求仿人多指手的抓取性能,相关研究采用全驱动和增加传感器的方式努力提高仿人多指手的灵活性,但是较高灵活性的同时也增加了结构的复杂度和制造成本。为了平衡仿人多指手的结构复杂度、操作灵活性与生产制造成本之间的矛盾关系,研究设计了一款欠驱动仿人多指手,并对其开展了抓取位置的规划分析。主要进行了以下方面的研究工作:设计了一款全新的仿人欠驱动多指手结构。该多指手为五指布局,总共11个主动自由度驱动17个自由度关节运动,其中拇指为全驱动3自由度结构;食指和中指为2驱动3自由度结构,其远指节和近指节采用耦合结构设计;无名指和小拇指则为2驱动4自由度结构。采用腱绳传动方式,其目的是为了提高手指内部空间利用率,为布置其他传感器提供了更多的空间,且腱绳布局更为灵活,更有利于从运动学角度开展手指整体设计。开展了仿人多指手运动学和动力学建模分析。基于D-H参数法推导了仿人多指手正运动学、逆运动学方程,得到手指关节转角与指尖位置的转换关系,进一步得到驱动腱绳位移量与关节转角变化量之间的映射关系,并基于matlab机器人工具箱求解了手指运动空间。基于雅可比矩阵,开展了仿人多指手静力学分析,得到手指自身重力、手指指尖负载与关节驱动力矩之间的转换关系,进一步求得驱动腱绳张力与关节驱动力矩之间的转换关系。基于运动学方程推导了仿人多指手动力学方程,得到了多指手关节驱动力矩与关节运动的映射关系。在运动学和动力学分析的基础上,开展了仿人多指手的抓取位置规划。根据抓取时接触手指数目与目标物体外观结构形状,给出了仿人多指手抓取位置规划的数学描述,通过分析接触条件将位置规划问题转化为优化问题。通过力封闭条件下的线性规划,获得多指手与物体接触的初始位置,进一步提出了位置抓取规划的稳定性指标,以最大稳定性指标为优化目标,对接触位置进行优化,并给出了具体优化算法流程。利用给出的优化算法流程开展了算例分析,取得了较好的优化效果。更多还原

【Abstract】 As the terminal operating mechanism of intelligent equipment,humanoid multi-fingered hand has the appearance structure and movement flexibility comparable to that of human hand,and can grasp many objects,which provides strong technical support and great convenience for human production activities and daily life.In order to pursue the grasping performance of humanoid multi-fingered hand,many universities and scientific research institutions strive to improve the flexibility of humanoid multi-fingered hand,but at the same time,there are some problems such as complex structure and high manufacturing cost.In order to balance the contradiction between the structural complexity,operational flexibility and manufacturing cost of humanoid multi-fingered hand,an underactuated humanoid multi-fingered hand was designed and its grasping position was planned and analyzed.The main research work is as follows:A new humanoid underactuated multi-fingered hand structure is designed.The multifingered hand has a five-finger layout,and 11 active degrees of freedom drive 17 degrees of freedom joint movements,in which the thumb is a fully driven 3-degree-of-freedom structure;The index finger and middle finger are 2-drive 3-DOF structures,and the distal and proximal knuckles are designed with coupling structures.The ring finger and the little finger are 2-drive4-DOF structures.Using the transmission mode of tendon-rope transmission,the structural design of key-driven finger is conducive to improving the utilization rate of finger internal space,providing more space for arranging other sensors,and the tendon-rope layout is more flexible,which is more conducive to analyzing the overall design of finger from the perspective of kinematics.The kinematics and dynamics analysis of humanoid multi-fingered hand is carried out.Based on D-H parameter method,the forward kinematics and inverse kinematics equations of humanoid multi-fingered hand are established,and the conversion relationship between finger joint angle and fingertip position is obtained,and the mapping relationship between the displacement of driving tendon and the change of joint angle is further obtained.The finger motion space is verified and solved based on matlab robot toolbox.Based on jacoby matrix,the static analysis of humanoid multi-fingered hand is carried out,and the conversion relationship between finger gravity,finger tip load and joint driving torque is obtained,and the conversion relationship between internal force of driving tendon and joint driving torque is further obtained.Based on the kinematics equation,the dynamic equation of humanoid multi-fingered hand is established,and the mapping relationship between joint driving torque and joint angle,acceleration and angular acceleration of multi-fingered hand is obtained.On the basis of kinematics and dynamics analysis,the grasping position planning of humanoid multi-fingered hand is carried out.According to the number of contact fingers and the appearance structure shape of the target object,the mathematical description of grasping position planning of humanoid multi-fingered hand is given,and the position planning problem is transformed into an optimization problem by analyzing the contact conditions.The initial contact position between the multi-fingered hand and the object satisfying the force closure condition is found by linear programming,and the stability index of position grasping planning is further proposed.Taking the maximum stability index as the optimization goal,the contact position is further optimized,and the specific optimization algorithm flow is given.An example analysis is carried out by using the given optimization algorithm flow,and good optimization results are obtained.更多还原