Robot Behavior Learning with a Dynamically Adaptive RBF Network : Experiments in Offline and Online Learning

In this paper a dynamically adaptive neural network architecture is investigated for robot behavior learning. Specifically, a so-called “Grow When Required” network (GWR) is used to dynamically cluster the sensor-motor training data for determining the centers of a radial basis function network (RBF), and then the RBF network is trained for acquiring and performing the required behaviors. We illustrated our learning system by making experiments from simple behaviors, such as wall following and obstacle avoiding, to some complicated behaviors, such as moving object following and path learning, which were conducted on a real robot. We also tested the learning system in online training mode. Experimental results showed that our learning system is able to learn a wide range of robot behaviors due to its dynamically adaptive learning structure.

[1]  Ulrich Nehmzow,et al.  Experiments in Competence Acquisition for Autonomous Mobile Robots , 1992 .

[2]  S. Hyakin,et al.  Neural Networks: A Comprehensive Foundation , 1994 .

[3]  Dean Pomerleau,et al.  ALVINN, an autonomous land vehicle in a neural network , 2015 .

[4]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[5]  Shinichi Nakasuka,et al.  Autonomous reconstruction of state space for learning of robot behavior , 2000, Proceedings. 2000 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2000) (Cat. No.00CH37113).

[6]  D. Saad,et al.  On-line learning of unrealizable tasks. , 1999, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[7]  Jun Tani,et al.  Articulation of sensory-motor experiences by forwarding forward model: from robot experiments to phenomenology , 2002 .

[8]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[9]  Óö,et al.  Learning Robot Behaviours with Self-Organizing Mapsand Radial Basis Fun tion Networks , .

[10]  David Saad.  The Theory of On-line Learning — A Statistical Physics Approach , 2003 .

[11]  Larry S. Davis,et al.  An improved radial basis function network for visual autonomous road following , 1996, IEEE Trans. Neural Networks.

[12]  Andres Perez-Uribe.  A non-computationally-intensive neurocontroller for autonomous mobile robot navigation , 2003 .

[13]  Nikola K. Kasabov,et al.  On-line learning, reasoning, rule extraction and aggregation in locally optimized evolving fuzzy neural networks , 2001, Neurocomputing.

[14]  Stefan Schaal,et al.  On-line Learning for Humanoid Robot Systems , 2000, ICML.

[15]  Ulrich Nehmzow,et al.  Route learning in mobile robots through self-organisation , 1996, Proceedings of the First Euromicro Workshop on Advanced Mobile Robots (EUROBOT '96).

[16]  Bernd Fritzke,et al.  Growing cell structures--A self-organizing network for unsupervised and supervised learning , 1994, Neural Networks.

[17]  Noel E. Sharkey,et al.  Learning from Innate Behaviors: A Quantitative Evaluation of Neural Network Controllers , 1998, Auton. Robots.

[18]  Stephen R. Marsland,et al.  A self-organising network that grows when required , 2002, Neural Networks.