Micro Assembly System For Automation Of The Research Of Control Methods Of The Flexible Manipulator Robot

As an advanced technology in 21 century,MEMS is becoming a new research field quickly.The traditional controlling method of force doesn't meet the demand of tiny force.The micro assembly system becomes the indispensable key of technology at the development process of MEMS,it doesn't gain satisfactory controlling effect with the controlling method of reliable force or location.The paper basing on research and exploration of literature at domestic and overseas study controlling and optimizing method aiming at construction of cantilever beam model,the traditional impedance controlling algorithm,the design and simulation of self-adaption impedance controlling algorithm,the regulation of impedance parameter on line with adding to fuzzy theory and fuzzy neural network combining with intelligence controlling algorithm,detailed information as follow:At first,the 3-D mobile platform and cantilever beam model are modeled dynamically,check effectiveness of fundamental impedance controlling algorithm by known model.The sports track and controlling force are accommodated into a dynamic frame according to equation of impedance controlling,so it avoids to need two controlling methods to control location and force and reduces controlling task,simplifies design of control system,at the same time it has high adaptable capacity to disturbance of outside and variation of environment.In view of difficulty of building precise dynamic model of mechanical arm within the allowable scope but doesn,t stabilize on a expected value.We design a self-adaption controlling algorithm that only needs an estimated model to attain expected value.And then due to all of inertia coefficient,damping coefficient and inflexibility coefficient in basis impedance controlling algorithm at impedance equations are set according to artificial experience and need abundant work volume to gain desired value.Introducing fuzzy theory can adjust impedance parameter automatically and controlling expected location and force can fulfill the task with saving time and being more precise.Finally,we present a controlling impedance method of using fuzzy neural network compensation,because there are kinds of effect of indeterminacy factors,the effect of environment model usually isn't precise and can't have foreknowledge.Simultaneously,wecome up with using wolf colony algorithm to optimize problems that BP algorithm of fuzzy neural network is easy to sink into part pole and slow convergence and gain expected result eventually.