Visually aided force control with fuzzy parameter tuning

Warning The system is temporarily closed to updates for reporting purpose.

Çallı, Berk (2008) Visually aided force control with fuzzy parameter tuning. [Thesis]

[thumbnail of BerkCalli.pdf] PDF

Download (1MB)


Vision and force sensors provide rich information which can enable robots to execute complex tasks. The integration of these two types of sensors may prove very useful in many industrial robotic applications, as well as for the robots that operate in environments where humans live. Vision sensors give robots the ability to operate in complex and dynamic environments. With force sensors contacts can be detected, and manipulation tasks can be done without the risk of damaging the workpiece. The integration of vision and force sensing systems equips robots with all these advantages and the abilities of robots can rise dramatically by the integrated use of these sensors. However, this integration is not straightforward. In this thesis, a literature survey about visual servoing and force control is presented firstly. Present integration methods are reported and discussed. A manipulation task is defined as a case study problem. In this problem, a constant magnitude normal force is to be exerted at a fixed point on an object which is free to rotate. Visual servoing and explicit force control techniques are applied next in the task frame formalism to achieve this objective. Disadvantages of the constant parameter controllers are addressed and two solutions in which controller gains are tuned with fuzzy logic systems are presented. The first solution is in the hybrid control category, whereas the second controller is a shared control strategy. These controllers are novel ones; they are the first applications of the fuzzy gain tuning on the integration of vision and force control systems. Experiments are carried out on a two degrees of freedom (DOF) direct drive SCARA type robot and the results obtained with fixed-parameter and fuzzy-tuned control methods are compared. The experimental results show that using fuzzy gain scheduling for the integration of force and vision systems improves the performance of combined controller and also prevents possible causes of instability.
Item Type: Thesis
Uncontrolled Keywords: Visual servoing. -- Force control. -- Sensor integration. -- Fuzzy logic. -- Online parameter tuning. -- Fuzzy control algorithms. -- Control algorithms. -- Görüntü tabanlı kontrol. -- Kuvvet kontrolü. -- Sensör birleştirme. -- Bulanık mantık. -- Çevrimiçi değiştirge ayarlama. -- Bulanık denetim algoritmaları. -- Denetim algoritmaları. -- Kuvvet denetimi. -- Görüntü tabanlı denetim.
Subjects: T Technology > TJ Mechanical engineering and machinery > TJ163.12 Mechatronics
Divisions: Faculty of Engineering and Natural Sciences > Academic programs > Mechatronics
Faculty of Engineering and Natural Sciences
Depositing User: IC-Cataloging
Date Deposited: 28 Jun 2010 16:49
Last Modified: 26 Apr 2022 09:51

Actions (login required)

View Item
View Item