| dc.description.abstract | Industrial manipulators have several applications in a multitude of disciplines. The use of industrial manipulators has increased rapidly, and they are more refined in many applications due to advances such as fast response time, high precision, quick speed and a high level of performance. Most industrial manipulators are position-controlled; usually vision and force sensors are not integrated in most commercial industrial robots. Therefore, the addition of force and vision sensing mechanisms is required to successfully automate advanced tasks, and to enable robots to avoid high contact forces while working in applications that require contact with environments.
The objective of this thesis is to implement a unilateral master-slave system for medical applications. In this thesis, a Polaris Vicra® optical tracking device is used to represent the master system, while a four degree of freedom (DOF) position-controlled SCARA manipulator from Epson is used to represent the slave system. The manipulator is equipped with a force-torque sensor to facilitate operation in unknown environments. In addition, MapleSim is used to find the dynamic model for the SCARA manipulator. Furthermore, MapleSim is also used to validate the control algorithm prior to implementation on the hardware.
Three force control techniques are used in this research and the robot's performance are evaluated. The control techniques are impedance control, admittance control and fuzzy logic control. The admittance and fuzzy logic controllers are applied to the proposed master-slave system while the impedance control is applied to the manipulator model, which was obtained from MapleSim.
In order to validate the presented control algorithms, several experiments and simulations were carried out. The experimental results show the ability of the presented controllers (admittance and fuzzy logic) to track the operator signal while keeping the force within the desired range. The simulation and animation of the impedance controller on the other hand, shows that the robot's performance can be evaluated through software. | en |
