Advertisement

Evolutionary Robotics Applied to Hexapod Locomotion: a Comparative Study of Simulation Techniques

  • Christiaan J. PretoriusEmail author
  • Mathys C. du Plessis
  • John W. Gonsalves
Article
  • 16 Downloads

Abstract

The Evolutionary Robotics (ER) process has been applied extensively to developing control programs to achieve locomotion in legged robots, as an automated alternative to the arduous task of manually creating control programs for such robots. The evolution of such controllers is typically performed in simulation by making use of a physics engine-based robotic simulator. Making use of such physics-based simulators does, however, have certain challenges associated with it, such as these simulators’ computational inefficiency, potential issues with lack of accuracy and the human effort required to construct such simulators. The current study therefore proposed and investigated an alternative method of simulation for a hexapod (six-legged) robot in the ER process, and directly compared this newly-proposed simulation method to traditional physics-based simulation. This alternative robotic simulator was built based solely on experimental data acquired directly from observing the behaviour of the robot. This data was used to construct a simulator for the robot based on Artificial Neural Networks (ANNs). To compare this novel simulation method to traditional physics simulation, the ANN-based simulators were used to evolve simple open-loop locomotion controllers for the robot in simulation. The real-world performance of these controllers was compared to that of controllers evolved in a more traditional physics-based simulator. The obtained results indicated that the use of ANN-based simulators produced controllers which could successfully perform the required locomotion task on the real-world robot. In addition, the controllers evolved using the ANN-based simulators allowed the real-world robot to move further than those evolved in the physics-based simulator and the ANN-based simulators were vastly more computationally efficient than the physics-based simulator. This study thus decisively indicated that ANN-based simulators offer a superior alternative to widely-used physics simulators in ER for the locomotion task considered.

Keywords

Hexapod Deep Learning Evolutionary Robotics Artificial Neural Networks Locomotion System identification 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Notes

References

  1. 1.
    Belter, D., Skrzypczynski, P.: A biologically inspired approach to feasible gait learning for a hexapod robot. Int. J. Appl. Math. Comput. Sci. 20(1), 69–84 (2010)CrossRefzbMATHGoogle Scholar
  2. 2.
    Bongard, J., Zykov, V., Lipson, H.: Resilient machines through continuous self-modeling. Science 314, 1118–1121 (2006)CrossRefGoogle Scholar
  3. 3.
    Bongard, J.C.: Evolutionary robotics. Commun. ACM 56(8), 74–83 (2013)CrossRefGoogle Scholar
  4. 4.
    Bullet Physics Library: Real-Time Physics Simulation (Accessed July 2016). http://bulletphysics.org/wordpress/
  5. 5.
    Chen, W., Ren, G., Zhang, J., Wang, J.: Smooth transition between different gaits of a hexapod robot via a central pattern generators algorithm. J. Intell. Robot. Syst. 67(3–4), 255–270 (2012)CrossRefzbMATHGoogle Scholar
  6. 6.
    Cook, D., Vardy, A.: Towards real-time robot simulation on uneven terrain using neural networks. In: 2017 International Joint Conference on Neural Networks (IJCNN), pp 1688–1695 (2017)Google Scholar
  7. 7.
    Corder, G., Foreman, D.: Nonparametric Statistics for Non-Statisticians: A Step-by-Step Approach. Wiley, New York (2011)zbMATHGoogle Scholar
  8. 8.
    Cully, A., Mouret, J.: Evolving a behavioral repertoire for a walking robot. Evol. Comput. 24(1), 59–88 (2016)CrossRefGoogle Scholar
  9. 9.
    Doncieux, S., Bredeche, N., Mouret, J.B., Eiben, A.E.G.: Evolutionary robotics: what, why, and where to. Front. Robot. AI 2, 4 (2015)CrossRefGoogle Scholar
  10. 10.
    Doncieux, S., Mouret, J.B.: Beyond black-box optimization: a review of selective pressures for Evolutionary Robotics. Evol. Intell. 7(2), 71–93 (2014)CrossRefGoogle Scholar
  11. 11.
    Drumwright, E., Hsu, J., Koenig, N., Shell, D.: Extending open dynamics engine for robotics simulation. Simulation, Modeling, and Programming for Autonomous Robots 6472, 38–50 (2010)CrossRefGoogle Scholar
  12. 12.
    Engelbrecht, A.P.: Computational Intelligence: An Introduction, 2nd edn. Wiley, West Sussex (2007)CrossRefGoogle Scholar
  13. 13.
    Farooq, S.S., Kim, K.J.: Evolution of neural controllers for simulated and real quadruped robots. In: Second International Conference on Robot, Vision and Signal Processing, pp 295–298. IEEE (2013)Google Scholar
  14. 14.
    Floreano, D., Mondada, F.: Evolution of homing navigation in a real mobile robot. IEEE Trans. Syst. Man Cybern. 26(3), 396–407 (1996)CrossRefGoogle Scholar
  15. 15.
    Glette, K., Klaus, G., Zagal, J.C., Torresen, J.: Evolution of locomotion in a simulated quadruped robot and transferral to reality. In: Proceedings of the Seventeenth International Symposium on Artificial Life and Robotics (2012)Google Scholar
  16. 16.
    Hartland, C., Bredeche, N.: Evolutionary robotics, anticipation and the reality gap. In: IEEE International Conference on Robotics and Biomimetics (2006)Google Scholar
  17. 17.
    Jakobi, N.: Running across the reality gap: octopod locomotion evolved in a minimal simulation. In: Evolutionary Robotics, pp 39–58. Springer (1998)Google Scholar
  18. 18.
    Jakobi, N., Husbands, P., Harvey, I.: Noise and the reality gap: the use of simulation in evolutionary robotics. In: Advances in Artificial Life, Lecture Notes in Computer Science, vol. 929, pp 704–720. Springer, Berlin (1995)Google Scholar
  19. 19.
    Jordan, M.I., Rumelhart, D.E.: Forward models: supervised learning with a distal teacher. Cogn. Sci. 16 (3), 307–354 (1992)CrossRefGoogle Scholar
  20. 20.
    Juang, C., Chang, Y., Hsiao, C.: Evolving gaits of a hexapod robot by recurrent neural networks with symbiotic species-based particle swarm optimization. IEEE Trans. Ind. Electron. 58(7), 3110–3119 (2011)CrossRefGoogle Scholar
  21. 21.
    Keras: The Python Deep Learning library (Accessed April 2017). https://keras.io/
  22. 22.
    Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. CoRR arXiv:1412.6980 (2014)
  23. 23.
    Klaus, G., Glette, K., Tørresen, J.: A comparison of sampling strategies for parameter estimation of a robot simulator. In: Proceedings of the Third International Conference on Simulation, Modeling, and Programming for Autonomous Robots (2012)Google Scholar
  24. 24.
    Koos, S., Cully, A., Mouret, J.: Fast damage recovery in robotics with the T-resilience algorithm. Int. J. Robot. Res. 32(14), 1700–1723 (2013)CrossRefGoogle Scholar
  25. 25.
    Koos, S., Mouret, J., Doncieux, S.: The transferability approach: crossing the reality gap in evolutionary robotics. IEEE Trans. Evol. Comput. 17, 1–25 (2012)Google Scholar
  26. 26.
    Moeckel, R., Perov, Y.N., Nguyen, A.T., Vespignani, M., Bonardi, S., Pouya, S., Sproewitz, A., van den Kieboom, J., Wilhelm, F., Ijspeert, A.J.: Gait optimization for roombots modular robots—matching simulation and reality. In: IEEE/RSJ International Conference on Intelligent Robots and Systems (2013)Google Scholar
  27. 27.
    Mouret, J.B., Chatzilygeroudis, K.: 20 years of reality gap: a few thoughts about simulators in evolutionary robotics. In: Proceedings of the Genetic and Evolutionary Computation Conference Companion, pp 1121–1124. ACM (2017)Google Scholar
  28. 28.
    Nelson, A.L., Barlow, G.J., Doitsidis, L.: Fitness functions in evolutionary robotics: a survey and analysis. Robot. Auton. Syst. 57(4), 345–370 (2009)CrossRefGoogle Scholar
  29. 29.
    OpenCV: Open Source Computer Vision (Accessed October 2016). http://opencv.org/
  30. 30.
    Parker, G.B.: The co-evolution of model parameters and control programs in Evolutionary Robotics. In: IEEE International Symposium on Computational Intelligence in Robotics and Automation, pp 162–167 (1999)Google Scholar
  31. 31.
    Pratihar, D.K.: Evolutionary robotics—a review. Sadhana 28(6), 999–1009 (2003)CrossRefGoogle Scholar
  32. 32.
    Pretorius, C.J., du Plessis, M.C., Cilliers, C.B.: Simulating robots without conventional physics: a neural network approach. J. Intell. Robot. Syst. 71(3–4), 319–348 (2013)CrossRefGoogle Scholar
  33. 33.
    Pretorius, C.J., du Plessis, M.C., Gonsalves, J.W.: A comparison of neural networks and physics models as motion simulators for simple robotic evolution. In: IEEE Congress on Evolutionary Computation (CEC) (2014)Google Scholar
  34. 34.
    Pretorius, C.J., du Plessis, M.C., Gonsalves, J.W.: Neuroevolution of inverted pendulum control: a comparative study of simulation techniques. J. Intell. Robot. Syst. 86(3), 419–445 (2017)CrossRefGoogle Scholar
  35. 35.
    Pretorius, C.J., du Plessis, M.C., Gonsalves, J.W.: The transferability of evolved hexapod locomotion controllers from simulation to real hardware. In: 2017 IEEE International Conference on Real-time Computing and Robotics (RCAR), pp 567–574 (2017)Google Scholar
  36. 36.
  37. 37.
    Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.: Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15(1), 1929–1958 (2014)MathSciNetzbMATHGoogle Scholar
  38. 38.
    TensorFlow: an open-source machine learning framework for everyone (Accessed April 2017). https://www.tensorflow.org/
  39. 39.
    Woodford, G.W., du Plessis, M.C.: Robotic snake simulation using ensembles of artificial neural networks in evolutionary robotics. In: Proceedings of the Genetic and Evolutionary Computation Conference, pp 173–180. ACM (2018)Google Scholar
  40. 40.
    Woodford, G.W., du Plessis, M.C., Pretorius, C.J.: Evolving snake robot controllers using artificial neural networks as an alternative to a physics-based simulator. In: 2015 IEEE Symposium Series on Computational Intelligence, pp 267–274. IEEE (2015)Google Scholar
  41. 41.
    Yosinski, J., Clune, J., Hidalgo, D., Nguyen, S., Zagal, J.C., Lipson, H.: Evolving robot gaits in hardware: the HyperNEAT generative encoding vs. parameter optimization. In: European Conference on Artificial Life, pp 890–897 (2011)Google Scholar
  42. 42.
    Zagal, J.C., Ruiz-del Solar, J.: Combining simulation and reality in Evolutionary Robotics. J. Intell. Robot. Syst. 50, 19–39 (2007)CrossRefGoogle Scholar

Copyright information

© Springer Nature B.V. 2019

Authors and Affiliations

  1. 1.Department of Mathematics and Applied MathematicsNelson Mandela UniversityPort ElizabethSouth Africa
  2. 2.Department of Computing SciencesNelson Mandela UniversityPort ElizabethSouth Africa

Personalised recommendations