Neurofuzzy Learning of Mobile Robot Behaviours

The era of mobile robotics for use in service and field applications is gaining momentum. The need for adaptability becomes self evident in allowing robots to evolve better behaviors to meet overall task criteria. We report the use of neuro-fuzzy learning for teaching mobile robot behaviors, selecting exemplar cases from a potential continuum of behaviors. Proximate active sensing was successfully achieved with infrared in contrast to the usual ultrasonics and viewed the front area of robot movement. The well-knowTi ANFIS architecture has been modified compressing layers to a necessary minimum with weight normalization achieved by using a sigmoidal function. Trapezoidal basis functions (B splines of order 2) with a partition of 1 were used to speed up computation. Reference to previous reinforcement learning results was made in terms of speed of learning and quality of behavior. Even with the limited input information, appropriate learning invariably took place in a reliable manner.

[1]  Nadine N. Tschichold-Gürman.  The neural network model RuleNet and its application to mobile robot navigation , 1997, Fuzzy Sets Syst..

[2]  Maja J. Mataric,et al.  Behaviour-based control: examples from navigation, learning, and group behaviour , 1997, J. Exp. Theor. Artif. Intell..

[3]  Andrew W. Moore,et al.  Reinforcement Learning: A Survey , 1996, J. Artif. Intell. Res..

[4]  Martin Brown,et al.  Neurofuzzy adaptive modelling and control , 1994 .

[5]  Maja J. Mataric,et al.  Studying the Role of Embodiment in Cognition , 1997, Cybern. Syst..

[6]  Andrew G. Barto,et al.  Reinforcement learning , 1998 .

[7]  Kai-Tai Song,et al.  Fuzzy-neuro control design for obstacle avoidance of a mobile robot , 1995, Proceedings of 1995 IEEE International Conference on Fuzzy Systems..

[8]  Maja J. Mataric,et al.  Reinforcement Learning in the Multi-Robot Domain , 1997, Auton. Robots.

[9]  Elias N. Houstis,et al.  Neurofuzzy Motion Planners for Intelligent Robots , 1997, J. Intell. Robotic Syst..

[10]  Mohan M. Trivedi,et al.  A neuro-fuzzy controller for mobile robot navigation and multirobot convoying , 1998, IEEE Trans. Syst. Man Cybern. Part B.

[11]  Hajime Asama,et al.  Simulation Environment For An Autonomous And Decentralized Multi-agent Robotic System , 1992, Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems.

[12]  Jacek M. Zurada,et al.  Computational Intelligence: Imitating Life , 1994 .

[13]  Richard S. Sutton,et al.  Introduction to Reinforcement Learning , 1998 .

[14]  I. D. Kelly,et al.  Increased learning rates through the sharing of experiences of multiple autonomous mobile robot agents , 1998, 1998 IEEE International Conference on Fuzzy Systems Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98CH36228).

[15]  M. Matarić.  Behavior-based robotics as a tool for synthesis of artificial behavior and analysis of natural behavior , 1998, Trends in Cognitive Sciences.

[16]  Richard S. Sutton,et al.  Reinforcement Learning , 1992, Handbook of Machine Learning.

[17]  John M. Evans,et al.  HelpMate: an autonomous mobile robot courier for hospitals , 1994, Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS'94).

[18]  Maja J. Mataric,et al.  Learning social behavior , 1997, Robotics Auton. Syst..

[19]  Chuen-Tsai Sun,et al.  Neuro-fuzzy And Soft Computing: A Computational Approach To Learning And Machine Intelligence [Books in Brief] , 1997, IEEE Transactions on Neural Networks.

[20]  François Charpillet,et al.  A new approach to design fuzzy controllers for mobile robots navigation , 1997, Proceedings 1997 IEEE International Symposium on Computational Intelligence in Robotics and Automation CIRA'97. 'Towards New Computational Principles for Robotics and Automation'.

[21]  Sorin Moga,et al.  From Perception-Action Loops to Imitation Processes: A Bottom-Up Approach of Learning by Imitation , 1998, Appl. Artif. Intell..