OPT OpenIR  > 研究生部
六自由度机械臂系统设计及其关键技术研究
师恒1,2
学位类型博士
导师刘朝晖
2017-11-28
学位授予单位中国科学院研究生院
学位授予地点北京
学位专业光学工程
关键词机械臂 Poe建模 运动学标定 U-k方程 滑模控制
摘要

本文建立了六自由度机械臂运动学与动力学模型。鉴于传统运动学建模法计算复杂且存在奇异解,采用基于旋量理论的POE建模法分别得到机械臂运动学方程和运动学误差模型。提出了一种结合POE公式、Lagrange法和U-K方程的动力学建模法,建立了受约束机械臂动力学的完整方程,该方程具有一定的通用性。

本文完成了六自由度串联机械臂系统的研究与设计。前三关节选用电机组合+蜗轮蜗杆的二级减速机构,后三关节采用直接驱动的结构形式,电控部分采用PMAC运动控制卡+伺服驱动器+直流电机组合的形式,调试完成了机械臂的单轴运动与6轴联动。在MATLAB中设计GUI界面求解得到机械臂的正逆运动学、正逆动力学和轨迹规划。实现MATLAB与Pro/E数据接口,分析了机械臂在实验环境中的工作空间。通过ANSYS分析了机械臂关键零部件在受力情况下的应变与应力,结果满足强度要求。

机械臂在约束情况下,末端会受到理想约束力与非理想约束力作用。考虑到非理想约束力的不确定性,提出了滑模控制法实现非理想约束力的跟踪控制。为了验证控制算法的有效性,针对末端受垂直约束的两关节机械臂进行仿真实验。考虑到受约束机械臂存在冗余变量,提出了一种基于U-K方程的动力学模型降阶法,并设计降阶滑模控制器来实现受约束机械臂末端轨迹和非理想约束力的跟踪控制。为了消弱滑模控制的抖振现象,提出了降阶自适应模糊滑模控制和降阶自适应神经网络滑模控制方法,实现了受约束机械臂高精度的控制与未知非理想约束力的逼近和补偿。实验结果表明:降阶自适应模糊滑模控制较单独采用滑模控制的控制精度提高了约10000倍,降阶自适应神经网络滑模控制较单独采用滑模控制的控制精度提高了约100000倍。

采用了基于旋量理论的POE建模法,得到了自主设计六自由度机械臂的运动学模型和运动学误差模型。提出了一种基于徕卡全站仪TC2003与BMR棱镜的直接测量法完成了机械臂运动学标定。进行了坐标系转化与重复定位精度的测定实验,重复定位精度最终测定为0.3mm,小于设计初提出的指标0.5mm,达到预期的指标要求。通过测量机械臂不同位姿的实际坐标值,采用最小二乘法辨识出运动学参数误差并进行误差补偿,将机械臂的平均绝对定位精度由设计初的6.11mm提高到了0.82mm。

其他摘要

The kinematics and dynamics modelings of the six-degree-of-freedom manipulator are established in this paper. In view of the complexity of the traditional kinematics modeling method and the existence of singular solution, the kinematic equation and the corresponding error model of the manipulator are obtained by the POE modeling method based on the spin theory, The dynamic modeling method is proposed by combining the POE formula, the Lagrange method and the U-K equation, and the complete expression of the constrained manipulator dynamic is established. The equation has certain versatility.

The system of the 6-DOF series manipulator is studied and designed in this article. The two deceleration mechanisms of first three joints adopt motor combination and worm gear and other three joints employ direct drive structure. The form of the electronic control parts includes PMAC motion control card, servo drive and DC motor combination. The single-axis motion and 6-axis linkage of the manipulator are completed. Designing the GUI interface in MATLAB, the positive and negative kinematics, positive and negative dynamics and trajectory planning of the manipulator are obtained. Realizing the data interface between MATLAB and Pro/E, the working space of the manipulator in the experimental environment is analyzed. The strain and stress of the key parts of the manipulator under the stress are analyzed by ANSYS, and the results meet the strength requirement.

The end-effecter will be subject to the ideal and non-ideal constraint forces effect in the case of constraints. Considering the uncertainty of non-ideal constraint force, the sliding mode control method is proposed to track and control the non-ideal constraint force. In order to verify the effectiveness of the control algorithm, the simulation experiment of the two-jointed manipulator with vertical constraint is put forward. Considering the redundant variables of the constrained manipulator, the dynamic model order reduction method based on the U-K equation is put forward. To track and control the end-effecter trajectory and the non-ideal constraint force, the corresponding sliding mode controller is designed. In order to weaken the chattering phenomenon in the sliding mode control, a reduced-order adaptive fuzzy sliding mode control and a reduced-order adaptive neural network sliding mode control method are proposed which can realize the high precision control and the approximation and compensation of the unknown non-ideal constraint force. The experimental results show that the reduced-order adaptive fuzzy sliding mode control is about 10000 times compared with that of sliding mode control alone and the reduced-order adaptive neural network sliding mode control is about 100000 times higher than that of sliding mode control alone.

The kinematic model and the kinematic error model of the 6-DOF manipulator are obtained by using the POE modeling method based on the spin theory. A direct measurement method based on the Leica total station TC2003 and BMR prisms is proposed to perform the kinematic calibration of the manipulator. The coordinate system transformation and repeat positioning accuracy are measured. The repeat positioning accuracy is 0.3mm, which is satisfied the desired index requirements of 0.5mm. The actual coordinate values of the manipulator in different positions are measured and the errors of the kinematic parameters are identified by the least squares method. The absolute positioning accuracy of the manipulator is increased from the initial 6.11mmto 0.82mm.

学科领域工业机器人技术
语种中文
文献类型学位论文
条目标识符http://ir.opt.ac.cn/handle/181661/29837
专题研究生部
作者单位1.中国科学院大学
2.中国科学院西安光学精密研究所
推荐引用方式
GB/T 7714
师恒. 六自由度机械臂系统设计及其关键技术研究[D]. 北京. 中国科学院研究生院,2017.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
师恒学位论文.pdf(3370KB)学位论文 开放获取CC BY-NC-SA请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[师恒]的文章
百度学术
百度学术中相似的文章
[师恒]的文章
必应学术
必应学术中相似的文章
[师恒]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。