Akademik Veri Yönetim Sistemi
Tezin Türü: Doktora
Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Mühendislik Fakültesi, Makina Mühendisliği Bölümü, Türkiye
Tezin Onay Tarihi: 2012
Öğrenci: ÖZGÜR BAŞER
Danışman: ERHAN İLHAN KONUKSEVEN
Özet:Haptic devices are evaluated in terms of transparency, stability, pose and force control accuracy. Precise velocity and acceleration estimations are needed in haptic devices for accurate simulations. This thesis firstly focuses on velocity and acceleration estimations from incremental encoders and proposes two new estimation techniques. The second goal in the thesis is transparency improvement. The transparency is a metric that shows how well a virtual model is reflected to the user. Conventional force control algorithms are not sufficient for high transparency. Therefore, as alternative, a motor current based torque feedback compensator is developed. The transparency improvement efforts by using only active actuator adversely affect stability in haptic devices. In order to improve both stability and transparency, a new hybrid control algorithm is developed by employing active actuator and passive brake. Pose and force control accuracy are mostly problematic in multi-DOF haptic manipulators. Since the deviation of any kinematic parameter in Jacobian matrix adversely affects the accuracy of pose and force control, kinematic calibration is indispensible for haptic manipulators. Besides, most of the haptic manipulators use capstan drive mechanisms in the joints due to their advantages; but cause transmission errors. Therefore, a theory is developed for the capstan drive transmission errors. Then, a closed-chain kinematic calibration technique including transmission errors is introduced for a 7 DOF haptic device. Adjustable posture ability of redundant haptic manipulators can be used for performance improvements. Finally, an optimal posture control algorithm is developed to improve the stability of the 7-DOF haptic device.