基于扩展任务的双臂机器人位姿优化  

Pose Optimization of Dual Manipulators System Based on Augmented Task

作　　者：郭鹏 房灵申[1,2] 赵明扬 朱思俊[1,3,2] 

机构地区：[1]中国科学院沈阳自动化研究所机器人学国家重点实验室,沈阳110016 [2]中国科学院大学,北京100049 [3]中国科学院沈阳自动化研究所扬州工程技术研究中心,江苏扬州225127

出　　处：《组合机床与自动化加工技术》2018年第1期8-12,17,共6页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：江苏省省政策引导类计划:激光拼焊焊缝跟踪与质量检测系统研制(BY201506301)

摘　　要：为了能够在不修改工业机器人控制器的逆解算法的前提下对双机械臂系统进行位姿优化,同时要求以双臂各向同性指标为强化目标,设计出基于扩展任务的双臂姿态优化方法。首先对双臂机器人的协作任务进行了运动学分析与建模,然后基于最近被提出的一种新型的通用冗余机械臂位姿优化方法对双臂位姿优化进行数学描述,并表明了优化指标-双臂各向同性指标的几何和物理意义。再利用蒙特卡洛采样方法和非线性回归拟合出了主机械臂在操作空间的位姿与协作任务的位姿之间的数学关系,即扩展任务的数学模型。最后在ABB Robot Studio机器人仿真软件中对双臂系统进行主从协作运动任务的仿真,结果表明:利用扩展任务的优化方法可以实现以各向同性指标强化为目标的双臂系统的位姿优化,可以使两只机械臂都远离肘关节奇点和肩关节奇点,并且表明本质上还可以减小机械臂的条件数的值。In order to optimize the pose of the dual manipulators system without modifying the inverse kinematics algorithms in the industrial robot controller. In the meantime the isotropic index of robot is enforced.A dual manipulators pose optimization Method based on augmented task is proposed. In this paper,firstly the kinematical analysis and modelling of the cooperative task of the dual manipulators are carried out,then the mathematic description of pose optimization is established based on a newpose optimization method for redundant manipulator. At last the relationship between the pose of the leader manipulator and the pose of the cooperative task is established based on Monte Carlo sampling and nonlinear regression method so the augmented task mathematical model is established. The leader-follower task simulation was performed in ABB Robot Studio,and the results shows that the pose optimization method can be realized and both manipulators are away from elbowsigularity and shoulder sigularity. In addition it shows that condition number of the manipulators is reduced.

关 键 词：位姿优化 双臂机器人 冗余机器人 

分 类 号：TH165[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]