Arc-length based Two-step Robot Motion Teaching Method for Dynamic Tasks

In this study, a new robot motion teaching method is proposed for dynamic robotic tasks. In the proposed teaching framework, the path geometry definition and timeparametrization process are separated. For the path geometry definition, waypoint-based teaching is used to secure motion accuracy and safety, and teaching-by-teleoperation is used to extract intuitive human motion easily. The direct motion dynamics transfer algorithm is developed to project the human motion dynamics into the pre-defined motion path according to the normalized arc-length of two different paths. The overall teaching procedure is empirically validated with a 6 DoF collaborative robot, and the result shows that the dynamic motion like pepper sprinkling can be easily taught by setting waypoints and swing handheld motion controller. This study suggests an intuitive and practical robot motion teaching method especially powerful for the dynamic robotic tasks.

[1]  Taewoo Kim,et al.  A robot teaching framework for a redundant dual arm manipulator with teleoperation from exoskeleton motion data , 2014, 2014 IEEE-RAS International Conference on Humanoid Robots.

[2]  Maya Cakmak,et al.  Trajectories and keyframes for kinesthetic teaching: A human-robot interaction perspective , 2012, 2012 7th ACM/IEEE International Conference on Human-Robot Interaction (HRI).

[3]  Wan Kyun Chung,et al.  Programming by Demonstration Using the Teleimpedance Control Scheme: Verification by an sEMG-Controlled Ball-Trapping Robot , 2019, IEEE Transactions on Industrial Informatics.

[4]  Petr Novák,et al.  HUMAN – ROBOT COLLABORATION IN INDUSTRY , 2016 .

[5]  Tamim Asfour,et al.  Unifying Representations and Large-Scale Whole-Body Motion Databases for Studying Human Motion , 2016, IEEE Transactions on Robotics.

[6]  SeungHoon Lee,et al.  The direct teaching and playback method for robotic deburring system using the adaptiveforce-control , 2009, 2009 IEEE International Symposium on Assembly and Manufacturing.

[7]  Darwin G. Caldwell,et al.  Imitation Learning of Positional and Force Skills Demonstrated via Kinesthetic Teaching and Haptic Input , 2011, Adv. Robotics.

[8]  Jun Morimoto,et al.  Skill learning and action recognition by arc-length dynamic movement primitives , 2018, Robotics Auton. Syst..

[9]  Lijun Zhao,et al.  Robot teaching by teleoperation based on visual interaction and extreme learning machine , 2018, Neurocomputing.

[10]  Norbert Krüger,et al.  A comparison of types of robot control for programming by Demonstration , 2016, 2016 11th ACM/IEEE International Conference on Human-Robot Interaction (HRI).

[11]  Satoru Goto,et al.  Human direct teaching of industrial articulated robot arms based on force-free control , 2006, Artificial Life and Robotics.

[12]  Aude Billard,et al.  Dynamical System Modulation for Robot Learning via Kinesthetic Demonstrations , 2008, IEEE Transactions on Robotics.

[13]  Pieter Abbeel,et al.  AVID: Learning Multi-Stage Tasks via Pixel-Level Translation of Human Videos , 2019, Robotics: Science and Systems.

[14]  Alessandro Gasparetto,et al.  Optimal trajectory planning for industrial robots , 2010, Adv. Eng. Softw..

[15]  Mariko Yamada,et al.  Development of teaching pendant optimized for robot application , 2009, 2009 IEEE Workshop on Advanced Robotics and its Social Impacts.