This thesis set out to determine whether a serial robot could be used to sand a wooden bowl, in order to free a human operator from what is considered a hazardous task. The process of sanding wood is similar to the polishing of metal, both set out to eliminate scratches. There are robot-based commercial systems available for the polishing of metal, for example aluminum, as employed by the aerospace industry. However, unlike aluminum, wood is a non-homogeneous material. In the case of wooden bowls, each has a unique geometry, and to a degree, unique material properties. The hypothesis posed by this thesis is that a hybrid force/position impedance controller can sand the inside of a wooden bowl and mimic the motions and technique of a human operator, and yet be able to deal with conditions of unknown bowl geometry. A CRS A465 robotic arm was used to mimic the arm motion of a human operator. The bowl was held in place by a vacuum chuck. A pneumatic orbital sander was used for the sanding process. A 3-axis force sensor mounted on the end effector measured the applied force. The arm was programmed to follow a radial line from the rim of the bowl to its center and then back to the rim. A hybrid force/position impedance controller was implemented. A forward and inverse kinematic model of the robot was developed to enable trajectory generation. The effect of sander angle and the nature of the trajectory was tested. PD action was adopted for the position controller. Four different force controllers were tested: First Order (FO) filter, Butterworth filter, PI action and combined FO/PI. Tuning tests were conducted to obtain the best values for the controller gains. The conclusion is that a FO filter for the force controller and PD action for the position controller, with an appropriate settings for the impedance, gave the best control performance. This performance was obtained with only a rough estimation of the bowl geometry as based on the bowl's diameter and depth. A precise CAD/CAM model of the bowl was not employed. iii ACKNOWLEDGEMENTS I would like to express my deepest gratitude to my supervisor, Dr. Brian Surgenor for his full support, expert guidance, understanding and incredible patience throughout my study and research. I have learned a lot from him throughout the completion of this thesis work. I would also like to thank my co-supervisor Prof. Keyvan Hashtrudi-Zaad and PdF. Chiedu Mokogwu for helping me to set up the CRS robot and force sensor. I would like to acknowledge the work of my friend: Hamza Sheikh. Thank you for your help with transferring inverse kinematic equations to the Simulink. I thank my lab mates Victor Luna, Andres Ramos, Shane Forbrigger and Orighomisan Mayuku for their support with moving equipment and testing of the robot. Thank you all for your helpful ideas, enthusiasm and friendship during the last two years. The financial support provided by the School of Graduate Studies and Department of Mechanical and Materials Engineering is greatly acknowledged. I must express my gratitude for my p...
