Skip to main content Accessibility help
×
Home

Complex Morphology Neural Network Simulation in Evolutionary Robotics

  • Grant W. Woodford (a1) and Mathys C. du Plessis (a1)

Summary

This paper investigates artificial neural network (ANN)-based simulators as an alternative to physics-based approaches for evolving controllers in simulation for a complex snake-like robot. Prior research has been limited to robots or controllers that are relatively simple. Benchmarks are performed in order to identify effective simulator topologies. Additionally, various controller evolution strategies are proposed, investigated and compared. Using ANN-based simulators for controller fitness estimation during controller evolution is demonstrated to be a viable approach for the high-dimensional problem specified in this work.

Copyright

Corresponding author

*Corresponding author. E-mail: s205014224@mandela.ac.za

References

Hide All
1.Bongard, J., “Evolutionary robotics”, Commun. ACM 56(8), 7483 (2013).
2.Nolfi, S. and Floreano, D., Evolutionary Robotics: The Biology, Intelligence, and Technology of Self-Organizing Machines (MIT Press, Cambridge, MA, 2000).
3.Woodford, G., du Plessis, M. and Pretorius, C., “Evolving Snake Robot Controllers Using Artificial Neural Networks as an Alternative to a Physics-Based Simulator”, 2015 IEEE Symposium Series on Computational Intelligence, Cape Town, South Africa (2015) pp. 267274.
4.Pretorius, C., du Plessis, M. and Cilliers, C., Simulating robots without conventional physics: A neural network approach, J. Intell. Robotic Syst., Springer 71(3–4), 319348(2013).
5.Pretorius, C., du Plessis, M. and Gonslaves, J., A Comparison of Neural Networks and Physics Models as Motion Simulators for Simple Robotic Evolution, 2014 IEEE Congress on Evolutionary Computation (CEC), Beijing, China (2014) pp. 27932800.
6.Prabhu, S., Seals, R., Kyberd, P., and Wetherall, J., A survey on evolutionary-aided design in robotics, Robotica 36(12), 18041821 (2018).
7.Floreano, D. and Mondada, F., Automatic creation of an autonomous agent: Genetic evolution of a neural-network driven robot, In: Cliff, D., Husbands, P., Meyer, J.-A., Wilson, S. (eds.), From Animals to Animats 3, (MIT Press, Cambridge, MA, 1994) pp. 421430.
8.Miglino, O., Lund, H. and Nolfi, S., Evolving mobile robots in simulated and real environments, Artificial Life 2(4), 417434 (1995).
9.Mouret, J. and Chatzilygeroudis, K. 20 Years of Reality Gap: A Few Thoughts About Simulators in Evolutionary Robotics, Proceedings of the Genetic and Evolutionary Computation Conference Companion, Berlin, Germany (2017) pp. 11211124.
10.Woodford, G., du Plessis, M. and Pretorius, C., Concurrent controller and Simulator Neural Network development for a snake-like robot in Evolutionary Robotics, Robot. Auton. Syst., 88, 3750 (2017).
11.Bongard, J., Zykov, V. and Lipson, H., Resilient machines through continuous self-modeling, Science, Elsevier 314(5802), 11181121 (2006).
12.Grefenstette, J. and Ramsey, C., An Approach to Anytime Learning, Mach. Learn. Proc. 189195 (1992).
13.Zagal, J. and Ruiz-Del-Solar, J., Combining simulation and reality in evolutionary robotics, J. Intell. Robot. Syst. 50(1), 1939 (2007).
14.Cully, A., Clune, J., Tarapore, D. and Mouret, J., Robots that can adapt like animals, Nature 521(7553), 503507 (2015).
15.Koos, S., Mouret, J. and Doncieux, S., The transferability approach: Crossing the reality gap in evolutionary robotics, Evol. Computat. 17(1), 122145 (2013).
16.Pretorius, C., du Plessis, M. and Gonslaves, J., Neuroevolution of Inverted Pendulum Control: A Comparative Study of Simulation Techniques, J. Intell. Robotic Syst. 86(3–4), 419445 (2017).
17.Nardi, R. and Holland, O., Coevolutionary modelling of a miniature rotorcraft, Intell. Auton. Syst. 10, 364 (2008).
18.Woodford, G., Pretorius, C. and du Plessis, M., Concurrent controller and Simulator Neural Network development for a differentially-steered robot in Evolutionary Robotics, Robot. Auton. Syst. 76, 8092 (2016).
19.Koopaee, J., Van Huijgevoort, B., Pretty, C. and Chen, X., Parameters tuning of snake robots sidewinding gait using Bayesian optimization, 2018 4th International Conference on Control, Automation and Robotics (ICCAR), Auckland, New Zealand (2018) pp. 4349.
20.Melo, K., Hernandez, M. and Gonzalez, D., Parameterized Space Conditions for the Definition of Locomotion Modes in Modular Snake Robots, 2012 IEEE International Conference on Robotics and Biomimetics (ROBIO), Guangzhou, China (2012) pp. 20322038.
21.Dowling, K., Limbless locomotion: Learning to crawl, Proceedings of the IEEE International Conference on Robotics and Automation, 1999, Detroit, Michigan, USA, vol. 4 (1999) pp. 30013006.
22.Kamegawa, T., Matsuno, F. and Chatterjee, R., Proposition of twisting mode of locomotion and GA based motion planning for transition of locomotion modes of 3-dimensional snake-like robot, Proceedings of the IEEE International Conference on Robotics and Automation, ICRA 2002, Washington, DC, USA, vol. 2 (2002) pp. 15071512.
23.Pretorius, C., du Plessis, M. and Gonsalves, J., Evolutionary robotics applied to hexapod locomotion: A comparative study of simulation techniques, J. Intell. Robotic Syst. 123 (2019).
24.Gal, Y. and Ghahramani, Z., Dropout as a Bayesian approximation: Insights and applications, Deep Learning Workshop, ICML (2015).
25.Melo, K., Paez, L. and Parra, C., Indoor and Outdoor Parametrized Gait Execution with Modular Snake Robots, IEEE International Conference on Robotics and Automation, St. Paul, MN, USA (2012).
26.Murata, S. and Kurokawa, H., Self-reconfigurable robots, Robot. Autom. Mag. 14(1), 7178 (2007).
27.Chollet, F., Keras, Github (2015. https://github.com/keras-team/keras
28.Martín, A. and Ashish, A., TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems, (2015). https://www.tensorflow.org
29.Kingma, D. and Ba, J., Adam: A method for stochastic optimization, (2014). arXiv preprint arXiv:1412.6980.

Keywords

Complex Morphology Neural Network Simulation in Evolutionary Robotics

  • Grant W. Woodford (a1) and Mathys C. du Plessis (a1)

Metrics

Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed