5487

5147

كارشناسي ارشد

طراحي كاربردي

اصفهان: دانشگاه صنعتي اصفهان، دانشكده مكانيك

1389

دوازده،112ص.: مصور،جدول،نمودار

ص.ع.به فارسي و انگليسي

مهدي كشميري

محمد دانش

4/8/89

محمد جعفر صديق،سعيد بهبهاني،فريد شيخ الاسلام

مهندسي مكانيك

ID5147

به فارسي و انگليسي: قابل رويت در نسخه ديجيتالي

An observer based state feedback controller design for robotic systems Mohamadreza Homayounzade m homayounzade@cc iut ac ir 1389 02 6 Department of Mechanical Engineering Isfahan University of Technology Isfahan 84156 83111 IranDegree M Sc Language Farsi Mehdi Keshmiri mehdik@cc iut ac irAbstract Over the last decade a lot of research effort has been put into design of sophisticated controlstrategies for robots with system partial state feedback However the poor quality of velocitymeasurements may significantly deteriorate the control performance of these methods In the area ofobserver design for nonlinear systems there is not one standard design like there is for linear systems In this thesis we propose two solutions for observer and observer based controller design Thefirst solution is applicable for serial robots but the second solution is extended the previous results to avast class of nonlinear systems including constrained and parallel robotic systems To this end first astable observer for system state estimation is developed and next the controller observer combination isconstructed such that the closed loop system be stable and finally the actuator dynamic is considered insystem equations The inclusion of actuators into the dynamic equations complicates the controllerstructure and its stability analysis since the systems are described by third order differential equations The designed controller consists of two parts observer based part that generates an estimated errorstate from the error on the joint position and a linear feedback part that utilizes this estimated state It isshown that this computationally efficient controller yields semi global asymptotic stability of thetracking error A key point in this design strategy is a fine tuning of the controller and observerstructure to each other which provides solutions to the output feedback robot control problem To show the validity of the proposed schemes the algorithm is implemented on a two link serialmanipulator and a slider crank as an example of constrained systems also the simulations are repeatedfor two case studies considering that they are actuated by BDC Brushed DC motors The proposedmethods are in fact model based but the simulations verify practically the robustness of the systemagainst system uncertainties and input disturbance

مهدي كشميري

محمد دانش

محمد جعفر صديق،سعيد بهبهاني،فريد شيخ الاسلام