Teaching a tele-robot using natural language commands

For Internet-based teleoperation systems, user-friendly natural interfaces are advantageous because those systems are intended to be used by non-experts. In developing user friendly interfaces, natural language communication is mandatory. This paper presents a system in which a sub-set of natural language is used to command a tele-robot manipulator doing an object sorting task. The paper discusses about referring to objects with natural language commands such as "pick the small red cube". This is achieved by learning individual lexical symbols that refer to colors, shapes, and sizes independently, and then inferring the meaning of a combination of them.

[1]  A. Ferworn,et al.  MAX: wireless teleoperation via the World Wide Web , 1999, Engineering Solutions for the Next Millennium. 1999 IEEE Canadian Conference on Electrical and Computer Engineering (Cat. No.99TH8411).

[2]  Ming-Kuei Hu,et al.  Visual pattern recognition by moment invariants , 1962, IRE Trans. Inf. Theory.

[3]  Francesco Mondada,et al.  KhepOnTheWeb: open access to a mobile robot on the Internet , 2000, IEEE Robotics Autom. Mag..

[4]  Kenneth Y. Goldberg,et al.  Desktop teleoperation via the World Wide Web , 1995, Proceedings of 1995 IEEE International Conference on Robotics and Automation.

[5]  Paul Vogt,et al.  The physical symbol grounding problem , 2002, Cognitive Systems Research.

[6]  Roque Marín,et al.  Including efficient object recognition capabilities in online robots: from a statistical to a Neural-network classifier , 2005, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[7]  Deb Roy,et al.  Mental imagery for a conversational robot , 2004, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[8]  Dana H. Ballard,et al.  A multimodal learning interface for grounding spoken language in sensory perceptions , 2004, ACM Trans. Appl. Percept..