Algoritmos Genéticos e Redes Neurais Artificiais no Controle de Robôs Móveis em Ambientes Dinâmicos

• Eder A. B. de Oliveira University of São Paulo
• Rafael Del Lama University of São Paulo
• Renato Tinós University of São Paulo

DOI: https://doi.org/10.5753/eniac.2023.234282

Abstract

The development of robust control laws for mobile robots in Dynamic Environments is a challenging problem. The objective of this work is the search for control solutions for mobile robots that adapt to changes in the environment. The proposed strategy starts from an algorithm developed by the programmer to perform a task in a predefined environment. Then, an Echo-State Network (ESN) is trained in a supervised way to reproduce the input-output mapping of the algorithm developed by the programmer. Finally, the trained ESN is adapted every time the environment changes. For this, the ESN weights are optimized using Genetic Algorithms (GAs). Evolutionary techniques developed for dynamic environments (GAs with random immigrants and hypermutation) are compared in experiments in which ten new environments are considered. The robustness of the control laws found is tested by presenting five new environments. In the simulations, the use of GAs allowed the adaptation of the control laws to the new environments. However, the solutions found were not necessarily robust to new changes in the environment.

Keywords: Genetic Algorithms, Evolutionary Robots, Dynamic Evolutionary Optimization

References

Aragón, V. S. and Esquivel, S. C. (2004). An evolutionary algorithm to track changes of optimum value locations in dynamic environments. Journal of Computer Science and Technology, 4(03):p. 127–133.

Beyer, H.-G. and Sendhoff, B. (2007). Robust optimization – a comprehensive survey. Computer Methods in Applied Mechanics and Engineering, 196:3190–3218.

Branke, J. (2001). Evolutionary Optimization in Dynamic Environments. Kluwer Academic Publishers, USA.

Cobb, H. G. and Grefenstette, J. J. (1993). Genetic algorithms for tracking changing environments. In Proceedings of the 5th International Conference on Genetic Algorithms, page 523–530, San Francisco, CA, USA. Morgan Kaufmann Publishers Inc.

Del Lama, R. S., Candido, R. M., Raineri, L. T., and Tinós, R. (2018). Evolutionary optimization of robust control laws for mobile robots in dynamic environments. In Anais do XV Encontro Nacional de Inteligência Artificial e Computacional, pages 461–472, Porto Alegre, RS, Brasil. SBC.

Floreano, D. and Nolfi, S. (2000). Evolutionary Robotics: The Biology, Intelligence, and Technology of Self-Organizing Machines. MIT Press/Bradford Books.

Fu, H., Sendhoff, B., K. Tang, K., and Yao, X. (2015). Robust optimization over time: Problem difficulties and benchmark problems. Evolutionary Computation, IEEE Transactions on, 19:731–745.

Goldberg, D. E. (1989). Genetic Algorithms in Search, Optimization and Machine Learning. Addison-Wesley Longman Publishing Co., Inc., USA, 1st edition.

Jaeger, H. (2001). The”echo state”approach to analysing and training recurrent neural networks-with an erratum note’. Bonn, Germany: German National Research Center for Information Technology GMD Technical Report, 148.

Jager, H. and Haas, H. (2004). Harnessing nonlinearity: Predicting chaotic systems and saving energy in wireless communication. Science, 304(5667):78–80.

Jin, Y. and Branke, J. (2005). Evolutionary optimization in uncertain environments-a survey. IEEE Transactions on Evolutionary Computation, 9(3):303–317.

Meyer, J. (1998). Evolutionary approaches to neural control in mobile robots. In SMC’98 Conference Proceedings. 1998 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.98CH36218), volume 3, pages 2418–2423 vol.3.

Mitchell, M. (1996). An Introduction to Genetic Algorithms. MIT Press/Bradford Books. Parasuraman, R. (2015). Few common failure cases in mobile robots. CoRR, abs/1508.03000.

Parasuraman, R. (2015). Few common failure cases in mobile robots. CoRR, abs/1508.03000.

Raineri, L., Lama, R. D., Candido, R., Costa, A., and Tinós, R. (2019). Robôs evolutivos para a investigação do comportamento de ratos no teste de campo aberto. Revista Eletrônica De Iniciação Científica Em Computação, 17(2).

Shimo, H. K., Roque, A. C., Tinós, R., Tejada, J., and Morato, S. (2010). Use of evolutionary robots as an auxiliary tool for developing behavioral models of rats in an elevated plus-maze. In Proc. of the 2010 Eleventh Brazilian Symposium on Neural Networks, pages 217–222. IEEE.

Webb, B. (2001). Can robots make good models of biological behaviour? Behavioral and Brain Sciences, 24(6):1033–1050.

Yu, X., Jin, Y., Tang, K., and Yao, X. (2010). Robust optimization over time — a new perspective on dynamic optimization problems. In IEEE Congress on Evolutionary Computation, pages 1–6.