An automatic vision guided position controller in a conveyor belt pick and place system

Thesis (Master)--Izmir Institute of Technology, Computer Engineering, Izmir, 2006Includes bibliographical references (leaves: 64-65)Text in English; Abstract: Turkish and Englishxii, 67 leavesAn automatic vision guided position controller system is developed as for possible applications such as handling and packaging that require position and orientation control. The aim here is to minimize the production cycle time, and to improve the economic performance and system productivity. The system designed can be partitioned into five major parts: vision module, pneumatic automation module, manipulator, conveyor-belt and a software that manages and integrates these modules. The developed software captures raw image data from a camera that is connected to a PC via usb port. Using image processing methods, this software determines the proper coordinates and pose of the moving parts on the conveyor belt in real time. The pick and place system locates the parts to the packaging area as part.s predefined orientation. The software communicates with a controller card via serial port, manages and synchronizes the peripherals (conveyor belt stepper motors- pneumatic valves,etc) of the system. C programming language is used in the implementation. OpenCV library is utilized for image acquisition. The system has the following characteristics: The Conveyor belt runs with a constant speed and objects on the conveyor belt may have arbitrary position and orientation. The vision system detects parts with their position and orientation on the moving conveyor belt based on a reference position. The manipulator picks the part and then corrects its position comparing the information obtained by vision system with predefined position, and it places the object to the packaging area. System can be trained for the desired position of the object