Modular Weightless Neural Network Architecture for Intelligent Navigation

The standard multi layer perceptron neural network (MLPNN) type has various drawbacks, one of which is training requires repeated presentation of training data, which often results in very long learning time. An alternative type of network, almost unique, is the Weightless Neural Network (WNNs) this is also called n-tuple networks or RAM based networks. In contrast to the weighted neural models, there are several one-shot learning algorithms for WNNs where training takes only one epoch. This paper describes WNNs for recognizes and classifies the environment in mobile robot using a simple microprocessor system. We use a look-up table to minimize the execution time, and that output stored into the robot RAM memory and becomes the current controller that drives the robot. This functionality is demonstrated on a mobile robot using a simple, 8 bit microcontroller with 512 bytes of RAM. The WNNs approach is code efficient only 500 bytes of source code, works well, and the robot was able to successfully recognize the obstacle in real time.

[1]  Michal Morciniec,et al.  A Theoretical and Experimental Account of n-Tuple Classifier Performance , 1996, Neural Computation.

[2]  I. Aleksander,et al.  WISARD·a radical step forward in image recognition , 1984 .

[3]  Paolo Coraggio,et al.  WiSARD and NSP for Robot Global Localization , 2007, IWINAC.

[4]  Wolfram Burgard,et al.  Probabilistic Robotics (Intelligent Robotics and Autonomous Agents) , 2005 .

[5]  E. do Valle Simoes,et al.  High speed neural control for robot navigation , 1996, 1996 IEEE International Conference on Systems, Man and Cybernetics. Information Intelligence and Systems (Cat. No.96CH35929).

[6]  Raida Al-Alawi.  FPGA Implementation of a Pyramidal Weightless Neural Networks Learning System , 2003, Int. J. Neural Syst..

[7]  Jim Austin.  A review of RAM based neural networks , 1994, Proceedings of the Fourth International Conference on Microelectronics for Neural Networks and Fuzzy Systems.

[8]  E. V. Simoes,et al.  An embedded evolutionary controller to navigate a population of autonomous robots , 2008 .

[9]  Michael C. Fairhurst,et al.  Adaptive pattern recognition using goal seeking neurons , 1991, Pattern Recognit. Lett..

[10]  Raida Al-Alawi,et al.  Performance Evaluation of Fuzzy Single Layer Weightless Neural Network , 2007, Int. J. Uncertain. Fuzziness Knowl. Based Syst..

[11]  W S McCulloch,et al.  A logical calculus of the ideas immanent in nervous activity , 1990, The Philosophy of Artificial Intelligence.

[12]  Igor Aleksander,et al.  Introduction to Neural Computing , 1990 .

[13]  Sivasubramaniam Ramanan,et al.  pRAM nets for detection of small targets in sequences of infra-red images , 1995, Neural Networks.

[14]  Dante Augusto Couto Barone,et al.  Hardware implementation of RAM neural networks , 1996, Pattern Recognit. Lett..

[15]  Igor Aleksander,et al.  General neural unit: retrieval performance , 1991 .

[16]  Ingemar J. Cox,et al.  Blanche-an experiment in guidance and navigation of an autonomous robot vehicle , 1991, IEEE Trans. Robotics Autom..

[17]  Teuvo Kohonen,et al.  An introduction to neural computing , 1988, Neural Networks.

[18]  Rosni Abdullah,et al.  Weightless Neural Network Array for Protein Classification , 2004, PDCAT.

[19]  John G. Taylor,et al.  Speaker identification for security systems using reinforcement-trained pRAM neural network architectures , 2001 .

[20]  Liqiang Feng,et al.  Navigating Mobile Robots: Systems and Techniques , 1996 .

[21]  Chandrasekhar Kambhampati,et al.  NEURAL NETWORK CONTROLLER FOR MOBILE ROBOT INSECT , 1994 .

[22]  Thomas Martini Jørgensen,et al.  Classification of Handwritten Digits using a RAM Neural Net Architecture , 1997, Int. J. Neural Syst..