Recognizing teleoperated manipulations

The many degrees-of-freedom and distributed sensing capability of dextrous robot hands permit the use of control programs that rely on qualitative changes in sensor feedback. One way of designing such a control program is to have the robot learn the qualitative control characteristics from examples. These examples may be provided via teleoperation. Results are presented for recognizing and segmenting manipulation primitives from a teleoperated task by analysis of features in sensor feedback. k-nearest quantized pattern vectors determine potential classifications. A hidden Markov model provides task context for the final segmentation. The illustrative task is picking up a plastic egg with a spatula.>