english only
EPFL > STI > IMT > LASA > Publications > Abstract



In robot Programming by Demonstration, current approaches for learning motion models rely on encoding the demonstrated trajectories using statistical techniques. In this paper we propose a method for extracting task constraints from demonstrated motions and using them directly as continuously embeddable constraints for controlling the robot. We consider determining the object of interest in each region of the task (frame of reference), and the contribution of the variable of interest, position vs. force on each axis. Furthermore the demonstrated motion can be segmented into meaningful segments based on the change of the task constraints.


Downloadable files: 0) { $tempFile = $row['pdfFile']; $temp = "pdf"; echo "[$temp] "; } // ps.Z if (strlen($row['psZFile'])>0) { $tempFile = $row['psZFile']; $temp = "ps.Z"; echo "[$temp] "; } // ps.gz if (strlen($row['psgzFile'])>0) { $tempFile = $row['psgzFile']; $temp = "ps.gz"; echo "[$temp] "; } ?>



Last update: 25/08/06