Adaptive Interface Mapping for Intuitive Teleoperation of Multi-DOF Robots

The supervisory control of multi-DOF robots by humans is a demanding application. If a single operator is tasked with direct control of a humanoid robot, performing coordinated tasks becomes non-intuitive and corresponds to unsustainable mental loads even for the most skilled operators. In this paper we use reinforcement learning to adaptively change the interface mapping from the operator user interface to the robot in such a way as to reduce the associated operator mental load. Based on the results of the interaction with the robot, we change the dynamical map describing the relationship between user commands and robot actions. The contribution of this paper is the adaptation of the interface map using reinforcement learning with reward functions associated with quantitative performance metrics. We present promising experimental results showing that the use of the proposed scheme can result in an easier to use interface map for a multi-DOF assistive robot controlled via a brain activity sensor.

[1]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[2]  Stefan Schaal,et al.  Reinforcement learning of motor skills in high dimensions: A path integral approach , 2010, 2010 IEEE International Conference on Robotics and Automation.

[3]  Jean Scholtz,et al.  Common metrics for human-robot interaction , 2006, HRI '06.

[4]  Peter Stone,et al.  Generalized model learning for Reinforcement Learning on a humanoid robot , 2010, 2010 IEEE International Conference on Robotics and Automation.

[5]  Dan O. Popa,et al.  Robot Head Motion Control with an Emphasis on Realism of Neck–Eye Coordination during Object Tracking , 2011, J. Intell. Robotic Syst..

[6]  Paolo Dario,et al.  Development of a Bioinstrumentation System in the Interaction between a Human and a Robot , 2006, 2006 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[7]  Michael A. Goodrich,et al.  Seven principles of efficient human robot interaction , 2003, SMC'03 Conference Proceedings. 2003 IEEE International Conference on Systems, Man and Cybernetics. Conference Theme - System Security and Assurance (Cat. No.03CH37483).

[8]  Gerhard Tröster,et al.  Discriminating Stress From Cognitive Load Using a Wearable EDA Device , 2010, IEEE Transactions on Information Technology in Biomedicine.

[9]  Iñaki Iturrate,et al.  Robot reinforcement learning using EEG-based reward signals , 2010, 2010 IEEE International Conference on Robotics and Automation.

[10]  Fillia Makedon,et al.  Neptune: assistive robotic system for children with motor impairments , 2011, PETRA '11.

[11]  James C. Maida,et al.  Improving Robotic Operator Performance Using Augmented Reality , 2007 .

[12]  Yael Edan,et al.  An objective function for performance measurement of human-robot target recognition systems in unstructured environments , 2004, 2004 IEEE International Conference on Systems, Man and Cybernetics (IEEE Cat. No.04CH37583).

[13]  Birsen Donmez,et al.  Evaluation criteria for human-automation performance metrics , 2008, PerMIS.

[14]  Jean Scholtz,et al.  Awareness in human-robot interactions , 2003, SMC'03 Conference Proceedings. 2003 IEEE International Conference on Systems, Man and Cybernetics. Conference Theme - System Security and Assurance (Cat. No.03CH37483).

[15]  Fakhri Karray,et al.  Towards generalized performance metrics for human-robot interaction , 2010, 2010 International Conference on Autonomous and Intelligent Systems, AIS 2010.