Applying Hybrid Reinforcement and Unsupervised Wieghtless Neural Network Learning Algorithm on Autonomous Mobile Robot Navigation.

Authors

  • Yusman Yusof Industrial Automation Section, Universiti Kuala Lumpur Malaysia
  • H.M. Asri H. Mansor France Institute, Bandar Baru Bangi, Selangor, Malaysia
  • H.M. Dani Baba Faculty of Electrical Engineering, Universiti Teknologi MARA, Shah Alam, Selangor, Malaysia

Keywords:

Reinforcement Learning, Q-learning, AutoWiSARD, Autonomous Navigation, Unsupervised Learning, Weightless Neural Network, LeJOS, Lego Mindstroms

Abstract

An autonomous system constructed using written computer programs based on human expert knowledge only handles anticipated and verified states. On the other hand, a self-learning algorithm allows an autonomous system to instinctively acquire knowledge, learn from experience and be more prepared to expect the unexpected. A novel hybrid selflearning algorithm which combines reinforcement and unsupervised weightless neural network algorithm learning was formulated. The self-learning algorithm was applied to an autonomous mobile robot navigation system in simulation and physical world. The result shows that the simulated and physical robot possesses the ability to self-learn by acquiring knowledge, learn and record experience without having prior information on the environment. The mobile robot was able to distinguish different types of obstacles i.e. corners and walls; and generate complex control sequences of actions in order to avoid these obstacles.

References

Iuri Wickert and Felipe M. G. França. 2001. AUTOWISARD:

Unsupervised Modes for the WISARD. 6th International WorkConference on Artificial and Natural Neural Networks, IWANN 2001. Granada, Spain. 13-15 June 2001. 435-441.

R. W. Yousif and M.A.H., Mansor. 2009. Design and Simulation of a New Self-Learning Expert System for Mobile Robot. International Journal of Computer, Electrical, Automation, Control and Information Engineering. 3(2).

U., Kamath U. 2008. Self-learning expert systems using rule classifier in detection engines. International Conference on Artificial Intelligence and Pattern Recognition, AIPR 2008, Orlando, Florida, USA. 7-10 July 2008. 224–227.

Y., Liu, Wang and M. Guo. 2005. The research and application of the self-learning expert system based on BP network. Fourth International Conference on Machine Learning and Cybernetics. 2005. China. 18-21 August 2005. 18–21.

C., Kirby, A., Sadlier, C.,Wood and M.,Vinther. 2013. Filling the Experience gap in the Drilling Optimization Continuous Improvement Cycle Through a Self-Learning Expert System. SPE Middle East Oil and Gas Show and Conference. 2013. Manama, Bahrain. 10-13 March 2013.

E., Piga and A., Geschiere. 2009 Self learning expert system (SLES) for power transformers. 20th International Conference and Exhibition on Electricity Distribution, CIRED 2009, Prague, Czech Republic. 2009. 8-11 June 2009. 1-3.

L., Chen and J., Li. 2012. Development and application of blast furnace expert system with self-learning function based on pattern recognition. Dongnan Daxue Xuebao (Ziran Kexue Ban)/Journal Southeast Univ. (Natural Sci. Ed.). 42(1):117–121.

D., Harel and A., Pnueli . 1985, Logics and Models of Concurrent Systems. New York: Springer-Verlag New York.

B., Mcelroy, M., Gillham, G., Howells, S., Spurgeon, S., Kelly, J., Batchelor and M.,Pepper. 2012. Highly efficient Localisation utilising Weightless neural systems. European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. 2012. Bruges, Belgium. 25-27 April 2012. 25–27.

S., Nurmaini . 2009. Modular Weightless Neural Network Architecture for Intelligent Navigation. International Journal of Soft Computing Application. 1(1): 1–18.

M., Gillham, B., McElroy, G., Howells, S., Kelly, S., Spurgeon and M., Pepper. 2012. Weightless Neural System Employing Simple Sensor Data for Efficient Real-Time Round-Corner, Junction and Doorway Detection for Autonomous System Path Planning in Smart Robotic Assisted Healthcare Wheelchairs. 2012 Third International Conference Emerging Security Technology. Lisbon, Portugal. 5-7 Sept. 2012. 161–164.

S. Nurmaini and B. Tutuko. 2011. A New Classification technique in Mobile Robot Navigation. TELKOMNIKA. 9(3): 453–464.

Building a Light-seeking Robot with Q-learning: InformIT retrieved October, 18, 2015 from

http://www.informit.com/articles/article.aspx?p=26423&seqNum=3.

Z., Wang, Z., Shi , Y., Li and J., Tu . 2013. The Optimization of Path Planning for Multi-robot System using Boltzmann Policy based QLearning Algorithm. 2013 IEEE International Conference on Robotics and Biomimetics (ROBIO). Shenzhen, China. 12-14 December 2013. 1199–1204.

H., Wicaksono. 2011. Q learning behavior on autonomous navigation of physical robot. 8th International Conference on Ubiquitous Robots and Ambient Intelligence (URAI). 2011. Incheon, Korea. 23-26 November 2011. 50–54.

H.,G.,A.,M., Víctor Ricardo Cruz-Álvarez and Enrique Hidalgo-Peña. 2012. A line follower robot implementation using Lego’s Mindstorms Kit and Q-Learning. Acta University. 22. 113–118.

S., Dini and M., Serrano. 2012. Combining Q-Learning with Artificial Neural Networks in an Adaptive Light Seeking Robot.

I., Aleksander and T.J., Stonham. 1979. Guide to pattern recognition using random access memories. IEE Journal Computing and Digital Technology. 2(1):29.

S., Sahin, M.R., Tolun and R., Hassanpour. 2012. Hybrid expert systems: A survey of current approaches and applications. Journal of Expert Systems and Applications. 39(4):4609–4617.

Simple Python robot simulator 2D download | SourceForge.net retrieved September, 14, 2015 from

http://sourceforge.net/projects/pyrobosim2d/.

Downloads

Published

2017-03-15

How to Cite

Yusof, Y., H. Mansor, H. A., & Dani Baba, H. (2017). Applying Hybrid Reinforcement and Unsupervised Wieghtless Neural Network Learning Algorithm on Autonomous Mobile Robot Navigation. Journal of Telecommunication, Electronic and Computer Engineering (JTEC), 9(1-3), 133–138. Retrieved from https://jtec.utem.edu.my/jtec/article/view/1758