Switchable task-priority framework for combining human-demonstrated and inverse kinematics tasks

We propose an approach that combines human demonstrated posture-control skill defined by the motion in the lower limb joints with an inverse kinematics solution of an arbitrary hand motion. The posture-control skill for humanoid robot was obtained through the human-in-the-loop teaching approach. The collected data during the teaching phase was used to approximate functional relation between the state of the robot's centre-of-pressure and the appropriate motion in the joints of the leg. The motion of the robot's hand was prescribed using the inverse kinematic solution. The posture-control skill and the motion of the hand were combined together using the task priority resolution strategy. We tested our approach with an experiment where the humanoid robot had to spin a spindle device with the hand and simultaneously maintain balance in the presence of external perturbations. In addition, we propose an upgrade that allows switching of control over the leg joints between arm task and posture-control task.

[1]  Stefano Chiaverini,et al.  Singularity-robust task-priority redundancy resolution for real-time kinematic control of robot manipulators , 1997, IEEE Trans. Robotics Autom..

[2]  Yoshikazu Kanamiya,et al.  Ankle and hip balance control strategies with transitions , 2010, 2010 IEEE International Conference on Robotics and Automation.

[3]  Jan Peters,et al.  Model learning for robot control: a survey , 2011, Cognitive Processing.

[4]  Tadej Petric,et al.  Teaching robots to cooperate with humans in dynamic manipulation tasks based on multi-modal human-in-the-loop approach , 2014, Auton. Robots.

[5]  Henk Nijmeijer,et al.  Robot Programming by Demonstration , 2010, SIMPAR.

[6]  F. Horak,et al.  Central programming of postural movements: adaptation to altered support-surface configurations. , 1986, Journal of neurophysiology.

[7]  Pierre-Brice Wieber,et al.  Analysis of the discontinuities in prioritized tasks-space control under discreet task scheduling operations , 2011, 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[8]  Stefan Schaal,et al.  Robot Programming by Demonstration , 2009, Springer Handbook of Robotics.

[9]  Matthias W. Seeger,et al.  Gaussian Processes For Machine Learning , 2004, Int. J. Neural Syst..

[10]  Andrej Gams,et al.  Reflexive stability control framework for humanoid robots , 2013, Auton. Robots.

[11]  Joshua G. Hale,et al.  Using Humanoid Robots to Study Human Behavior , 2000, IEEE Intell. Syst..

[12]  Nikolaos G. Tsagarakis,et al.  Tele-impedance: Teleoperation with impedance regulation using a body–machine interface , 2012, Int. J. Robotics Res..

[13]  Luka Peternel,et al.  Humanoid robot posture-control learning in real-time based on human sensorimotor learning ability , 2013, 2013 IEEE International Conference on Robotics and Automation.

[14]  Twan Koolen,et al.  Capturability-based analysis and control of legged locomotion, Part 1: Theory and application to three simple gait models , 2011, Int. J. Robotics Res..

[15]  Yoshihiko Nakamura,et al.  Inverse kinematic solutions with singularity robustness for robot manipulator control , 1986 .

[16]  A. A. Maciejewski,et al.  Obstacle Avoidance , 2005 .

[17]  Luka Peternel,et al.  Learning of compliant human–robot interaction using full-body haptic interface , 2013, Adv. Robotics.

[18]  Eiichi Yoshida,et al.  Humanoids , 2016, Springer Handbook of Robotics, 2nd Ed..

[19]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[20]  Jan Peters,et al.  Nonamemanuscript No. (will be inserted by the editor) Reinforcement Learning to Adjust Parametrized Motor Primitives to , 2011 .