The use of robots in industrial applications has increased significantly in recent years. Robots are faster, more precise, and stronger than humans, and they are capable of operating continuously without interruption. Nevertheless, humans still outperform robots in certain object manipulation tasks. One such task is object pushing. Humans perform this task almost intuitively by continuously observing the object and making small adjustments to the applied force and pushing direction, allowing the object to be moved accurately to the desired location. Since robots do not possess such sensory capabilities, alternative methods must be developed to enable reliable object pushing.
This bachelor's thesis presents a solution to this problem based on closed-loop control. By integrating a camera with a robotic manipulator, an algorithm was developed to guide the robot in pushing an object to a predefined target location. The thesis describes the complete development process, including communication with the robot, image acquisition and processing, and the challenges encountered during the development of the control algorithm. Finally, a series of experiments was conducted to evaluate the effectiveness and positioning accuracy of the proposed object-pushing approach.
|
