Autonomous Robot Navigation using Genetic Algorithms

Robots are often desired for tasks in hazardous environments. The robot may need to navigate in obstacle-filled, or uncertain, areas. Thus, the robot needs to be able to determine a feasible path for navigation. Due to the complexity of the path-planning problem, heuristic optimization methods such as genetic algorithms are often used. However, the models used by these algorithms can also be complex, which results in slow processing time.

Our group is currently working on improving the path representation model, so that it can be more efficiently processed. We have developed various models which we have tested on simulated navigation environments. The results of our work have also demonstrated a need for standardization in classification of obstacle-filled environments. This information is helpful in the development and testing of robot navigation algorithms, especially when comparing the performances of various methods. Thus, we are also developing a set of standard navigation environments, along with methods to classify them in terms of obstacle complexity.

Group Members

Faculty

Collaborators

Dr. Daniel Ashlock, University of Guelph
Dr. Jay Hanan, Oklahoma State University

Former Students and Theses

Andrew Hand, "A Tool for Benchmarking Robot Path Planning", M.S. Thesis, 2007.
Jagruthi Godugu, "Development of a Benchmark for Robot Path Planning", M.S. Thesis, 2004.
Kamran H-Sedighi, "Local Path Planning of an Autonomous Mobile Robot Using a Genetic Algorithm", M.S. Thesis, 2003.
Aditia Hermanu, "Genetic Algorithm with Modified Novel Value Encoding Technique for Autonomous Robot Navigation", M.S. Thesis, 2002.
Thomas Geisler, "Autonomous Robot Navigation System using a Genetic Algorithm with a Novel Value Encoding Technique", M.S. Thesis, 2002. First place award, 5th Annual TU Student Research Colloquium.

Publications

"Genetic Algorithms for Autonomous Robot Navigation", T.W. Manikas, K. Ashenayi, R.L. Wainwright, IEEE Instrumentation & Measurement Magazine, vol. 10, no. 6, Dec. 2007, pp. 26-31. [Invited Paper] (PDF)
"Evolving a Diverse Collection of Robot Path Planning Problems", D.A. Ashlock, T.W. Manikas, and K. Ashenayi, Proc. 2006 IEEE Congress on Evolutionary Computation (CEC2006), p. 1837-1844 (PDF)
"Benchmarking of Robot Path Planning Algorithms", A. Hand, J. Godugu, K. Ashenayi, T.W. Manikas, and R.L. Wainwright, in Intelligent Engineering Systems Through Artificial Neural Networks: Smart Engineering Systems Design: Neural Networks, Fuzzy Logic, Evolutionary Programming, Complex Systems and Artificial Life, C.H. Dagli, et al., Editors. 2005, ASME Press: New York. (PDF)
“Autonomous Robot Navigation Using a Genetic Algorithm with an Efficient Genotype Structure”, A. Hermanu, T.W. Manikas, K. Ashenayi, and R.L. Wainwright, in Intelligent Engineering Systems Through Artificial Neural Networks: Smart Engineering Systems Design: Neural Networks, Fuzzy Logic, Evolutionary Programming, Complex Systems and Artificial Life, C.H. Dagli, et al., Editors. 2004, ASME Press: New York. p. 319-324. (PDF)
"Autonomous Local Path Planning for a Mobile Robot Using a Genetic Algorithm", K. H-Sedighi, K. Ashenayi, T.W. Manikas, R.L. Wainwright, H.M. Tai, Proc. 2004 IEEE Congress on Evolutionary Computation (CEC2004), p. 1338-1345. (PDF)
"Development of a Genetic Algorithm Based Path Planner", K. H-Sedighi, Proc. 78th Annual AAAS - SWARM Conf., 2003.
"Development of a Benchmark for Robot Path Planning", J. Godugu, Proc. 78th Annual AAAS - SWARM Conf., 2003.
"Autonomous Robot Navigation System Using a Novel Value Encoded Genetic Algorithm", T. Geisler and T.W. Manikas, Proc. 45th IEEE Int. Midwest Symp. on Circuits and Systems, 2002, p. 45-48. (PDF)

This page last updated 2008 May 15