Vision application of human robot interaction: development of a ping pong playing robotic arm

Kalpesh Prakash Modi

Note: imported from RIT’s Digital Media Library running on DSpace to RIT Scholar Works in December 2013. Physical copy available through RIT's The Wallace Library at: QA76.9.H85 M63 2005


Robotics is a science that is implemented parallel to human behavior. This work describes and implements techniques to mathematically model the game of ping pong played by the humans, and utilization of these methods in the design and development of a ping pong playing robotic arm as an application of robotic vision. Displaced frame difference (DFD) is used to segment the ball motion from background motion and parametric calibration of single CCD camera is utilized to track the ball in three dimensions. This visual information is temporally updated and further applied to guide a robot arm to hit the ball at a specified location in time. The results signify the system development based on single camera tracking and also demonstrate its working with selfsufficiency for the color of the ball. System latency is measured as a function of the camera interface, processor architecture, and robot motion. Various hardware and software parameters that influence the real time system performance are also discussed. Keywords: Real time machine vision, camera calibration, 3-D imaging, robot vision.