Robotic Arm Control for Dice Manipulation using Computer Vision and Tactile Sensing

In the pursuit of advanced robotics and intelligent machine capabilities, this abstract explores the solutions adopted in a robotic challenge. Given a specific number as the goal, the objective entailed manipulating the dice to precisely exhibit the face corresponding to that specified value. The approach focused on four key aspects: design, control, vision, and perception. A gripper structure was designed and 3D-printed, control strategies were implemented to achieve the goal, a calibrated camera was utilized for dice recognition and value extraction, and sensors were integrated onto the gripper fingers for precise feedback during manipulation. This abstract explores the methodologies applied for each of these aspects and presents results and conclusions, where it is shown how the goal is achieved minimizing the number of iterations and the total time required.