Robotic Catching of Thrown Objects through a Monocular Vision System

Catching a thrown object through robotic systems requires several capabilities like smart sensing, object tracking, motion prediction, on-line trajectory planning and motion coordination. In the robotic literature, several papers deal with such a problem as well as the problem of motion trajectory estimation. Most of the approaches use either a stereo vision system to solve the 3D catching problem or a single camera for the 2D case. This scenario is reasonable because 3D tracking of the ball takes benefits from triangulation methods while, in the case of a single camera, only 2D information is directly available. However, high frame rate and optics with reasonable accuracy are required to achieve an accurate and fast trajectory prediction, i.e., a successful catch.

By using only one camera, the cost of the equipment can be reduced. Moreover, the calibration procedure for one camera is easier than in the stereo case. Hence, Fabio Ruggiero is co-author of an approach in which only one camera in eye-in-hand configuration is employed here, here, and here. A visual algorithm based on either the ball color or the circular shape, a new on-line trajectory estimation algorithm (a simple parabolic motion is assumed here, while the air drag effect is included here, here, and here ) and a partitioned-based visual approach are the principal components of the proposed solution. In particular, the solution here deals also with rolling and bouncing balls.