机器人关节动力学参数模糊辨识方法的研究-机械电子工程专业论文.docxVIP

摘 摘 要 机器人等机电系统关节面对整体结构系统动态特性有重要影响，辨 识出关节面的等效连接参数对于研究整机结构动态特性和动态控制有重 要价值。本论文以德国Amtex公司生产的九自由度模块化冗余机器人 (POWER CUBE)为研究对象，研究了采用模糊数学理论辨识机器人关节 面参数的方法。 论文对九自由度模块化机器人在垂直位姿及机器人臂绕第五个关节 逆时针旋转90度两个位姿分别进行了整机实验模态分析，得到了系统在 相应位姿下的固有频率、固有振型等模态参数，了解了机器人系统的动 态特性。另外，对绕第五个关节旋转45度和1 35度两个位姿分别测试了 关键点间的传递函数。 利用有限元法建立了机器人各模块的有限元模型，采用有限元法与 实验模态分析相结合的混和动力学建模方法，建立了机器人结构系统动 力学模型。 本文提出了基于模糊数学理论辨识机器人关节面参数的方法，采用 该方法，利用模块化机器人有限元模型和整机实验测试结果，辨识出了 九自由度模块化机器人的关节面参数。由此建立的机器人结构系统动力 学模型的动态特性与实验测试结果具有较好的一致性，表明所提出的方 法是有效的。另外，本文还提出了基于传递函数测试的关节面参数模糊 辨识方法，通过算例分析，与一般的基于传递函数测试的关节面参数辨 识方法进行了对比，验证了所提方法的有效性。 关键词：机器人关节面参数辨识模糊优化 ABSTRACTThe ABSTRACT The ioints of the electro．mechanical system have quite important effects on the dynamic characteristics of the whole structure．It is of quite valuable to identifv the ioint parameters to the research of the dynamic characteristics of the whole structure and its dynamic contr01．The robot studied in this thesis iS a 9-degrees of freedom roodular robot named POWER CUBE．which iS made in German Amtex Corporation． nle methods of identi母ing the joint parameters of the robot using fuzzy theory are studied． Experimental modal analysis to the modular robot in the vertical gesture and in the gesture that the robot tunis 90 degrees around the fifth joint has been done respectively The natural frequencies and mode shapes of the robot in the two gestures are obtained and the dynamic characteristics of the robot system are studied．The transfer functions between the key points are measured in the gestures that the robot turns 45 degrees and 135 degrees around the fi绌ioim respecfively_ The modules of the robot are analyzed with finite element method，the whole dynamic model ofthe robot carl be obtained via combining the finite element models of the modules with experimental modal analytical model． The method applying fuzzy theory tO identify the joint parameters is proposed in this Paper．Witll this method，the ioint parameters of the 9 DOFs modular robot are identified by applying the fir