New memory model for humanoid robots - introduction of co-associative memory using mutually coupled chaotic neural networks

Personal robots, which are expected to become popular in the future, are required to be active in joint work and community life with humans. Therefore, we have been developing new mechanisms and functions for a humanoid robot that has the ability to express emotions and to communicate with humans in a human-like manner. In 2004, we introduced the "Behavior Model" and "Consciousness Model" to the robot mental model so that the robot generated various kinds of behavior and an object of the robot's behavior became clear. We implemented the mental model in the emotion expression humanoid robot WE-4RII (Waseda Eye No.4 Refined II). Also, we have been studying a system of multiple harmonic oscillators (neurons) interacting via chaotic force since 2002. Each harmonic oscillator is driven by chaotic force whose bifurcation parameter is modulated by the position of the harmonic oscillator. In this paper, we propose an associative memory model using mutually coupled chaotic neural networks for generating an optimum behavior to a stimulus. We implemented this model in the emotional expression humanoid robot WE-4RII (Waseda Eye No.4 Refined II).

[1]  Paolo Dario,et al.  Effective emotional expressions with expression humanoid robot WE-4RII: integration of humanoid robot hand RCH-1 , 2004, 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566).

[2]  Brian Scassellati,et al.  How to build robots that make friends and influence people , 1999, Proceedings 1999 IEEE/RSJ International Conference on Intelligent Robots and Systems. Human and Environment Friendly Robots with High Intelligence and Emotional Quotients (Cat. No.99CH36289).

[3]  J. J. Hopfield,et al.  “Neural” computation of decisions in optimization problems , 1985, Biological Cybernetics.

[4]  Tatsuzo Ishida,et al.  Mechanical system of a small biped entertainment robot , 2003, Proceedings 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2003) (Cat. No.03CH37453).

[5]  Toshihiro Shimizu,et al.  Relaxation and bifurcation in Brownian motion driven by a chaotic force , 1990 .

[6]  Toshihiro Shimizu,et al.  Chaotic Brownian network , 1998 .

[7]  Paolo Dario,et al.  Design, fabrication and preliminary results of a novel anthropomorphic hand for humanoid robotics: RCH-1 , 2004, 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566).

[8]  Masahiro Fujita,et al.  Autonomous behavior control architecture of entertainment humanoid robot SDR-4X , 2003, Proceedings 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2003) (Cat. No.03CH37453).

[9]  Toshihiro Shimizu,et al.  The virtual attractor in mutually coupled networks , 2002 .

[10]  Atsuo Takanishi,et al.  Introduction of the need model for humanoid robots to generate active behavior , 2003, Proceedings 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2003) (Cat. No.03CH37453).

[11]  J J Hopfield,et al.  Neurons with graded response have collective computational properties like those of two-state neurons. , 1984, Proceedings of the National Academy of Sciences of the United States of America.

[12]  Atsuo Takanishi,et al.  Design and control of 9-DOFs emotion expression humanoid arm , 2004, IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004.

[13]  Masahiro Fujita,et al.  Behavior selection and motion modulation in emotionally grounded architecture for QRIO SDR-4XII , 2004, 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566).

[14]  Toshihiro Shimizu,et al.  New type of feed-back mechanism with bifurcation parameter modulation , 2001 .