Evolving Robots on Easy Mode: Towards a Variable Complexity Controller for Quadrupeds

  • Tønnes F. NygaardEmail author
  • Charles P. Martin
  • Jim Torresen
  • Kyrre Glette
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11454)


The complexity of a legged robot’s environment or task can inform how specialised its gait must be to ensure success. Evolving specialised robotic gaits demands many evaluations—acceptable for computer simulations, but not for physical robots. For some tasks, a more general gait, with lower optimization costs, could be satisfactory. In this paper, we introduce a new type of gait controller where complexity can be set by a single parameter, using a dynamic genotype-phenotype mapping. Low controller complexity leads to conservative gaits, while higher complexity allows more sophistication and high performance for demanding tasks, at the cost of optimization effort. We investigate the new controller on a virtual robot in simulations and do preliminary testing on a real-world robot. We show that having variable complexity allows us to adapt to different optimization budgets. With a high evaluation budget in simulation, a complex controller performs best. Moreover, real-world evolution with a limited evaluation budget indicates that a lower gait complexity is preferable for a relatively simple environment.


Evolutionary robotics Real-world evolution Legged robots 


  1. 1.
    Mouret, J., Doncieux, S.: Overcoming the bootstrap problem in evolutionary robotics using behavioral diversity. In: IEEE Congress on Evolutionary Computation, pp. 1161–1168, May 2009Google Scholar
  2. 2.
    Nygaard, T.F., Martin, C.P., Samuelsen, E., Torresen, J., Glette, K.: Real-world evolution adapts robot morphology and control to hardware limitations. In: Proceedings of the Genetic and Evolutionary Computation Conference. ACM (2018)Google Scholar
  3. 3.
    Nygaard, T.F., Martin, C.P., Torresen, J., Glette, K.: Self-modifying morphology experiments with DyRET: dynamic robot for embodied testing. In: IEEE International Conference on Robotics and Automation (ICRA), May 2019Google Scholar
  4. 4.
    Auerbach, J.E., Bongard, J.C.: Environmental influence on the evolution of morphological complexity in machines. PLoS Comput. Biol. 10(1), e1003399 (2014)CrossRefGoogle Scholar
  5. 5.
    Gong, D., Yan, J., Zuo, G.: A review of gait optimization based on evolutionary computation. Appl. Comput. Intell. Soft Comput. 2010, 12 (2010)CrossRefGoogle Scholar
  6. 6.
    Golubovic, D., Hu, H.: GA-based gait generation of Sony quadruped robots. In: Proceedings of the 3rd IASTED International Conference on Artificial Intelligence and Applications (AIA) (2003)Google Scholar
  7. 7.
    Moore, J.M., McKinley, P.K.: A comparison of multiobjective algorithms in evolving quadrupedal gaits. In: Tuci, E., Giagkos, A., Wilson, M., Hallam, J. (eds.) SAB 2016. LNCS (LNAI), vol. 9825, pp. 157–169. Springer, Cham (2016). Scholar
  8. 8.
    Hebbel, M., Nistico, W., Fisseler, D.: Learning in a high dimensional space: fast omnidirectional quadrupedal locomotion. In: Lakemeyer, G., Sklar, E., Sorrenti, D.G., Takahashi, T. (eds.) RoboCup 2006. LNCS (LNAI), vol. 4434, pp. 314–321. Springer, Heidelberg (2007). Scholar
  9. 9.
    Seo, K., Hyun, S., Goodman, E.D.: Genetic programming-based automatic gait generation in joint space for a quadruped robot. Adv. Robot. 24(15), 2199–2214 (2010)CrossRefGoogle Scholar
  10. 10.
    Pugh, J.K., Soros, L.B., Szerlip, P.A., Stanley, K.O.: Confronting the challenge of quality diversity. In: Proceedings of the 2015 Annual Conference on Genetic and Evolutionary Computation, pp. 967–974. ACM (2015)Google Scholar
  11. 11.
    Cully, A., Clune, J., Tarapore, D., Mouret, J.B.: Robots that can adapt like animals. Nature 521(7553), 503 (2015)CrossRefGoogle Scholar
  12. 12.
    de Santos, P.G., Garcia, E., Estremera, J.: Quadrupedal Locomotion: An Introduction to the Control of Four-Legged Robots. Springer, London (2007). Scholar
  13. 13.
    Nygaard, T.F., Samuelsen, E., Glette, K.: Overcoming initial convergence in multi-objective evolution of robot control and morphology using a two-phase approach. In: Squillero, G., Sim, K. (eds.) EvoApplications 2017. LNCS, vol. 10199, pp. 825–836. Springer, Cham (2017). Scholar
  14. 14.
    Nygaard, T.F., Torresen, J., Glette, K.: Multi-objective evolution of fast and stable gaits on a physical quadruped robotic platform. In: IEEE Symposium Series on Computational Intelligence (SSCI), December 2016Google Scholar
  15. 15.
    Ijspeert, A.J.: Central pattern generators for locomotion control in animals and robots: a review. Neural Netw. 21(4), 642–653 (2008)CrossRefGoogle Scholar
  16. 16.
    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: ECAL, pp. 890–897 (2011)Google Scholar
  17. 17.
    Togelius, J.: Evolution of a subsumption architecture neurocontroller. J. Intell. Fuzzy Syst. 15(1), 15–20 (2004)Google Scholar
  18. 18.
    Tomko, N., Harvey, I.: Do not disturb: recommendations for incremental evolution. In: Proceedings of ALIFE XII, the 12th International Conference on the Synthesis and Simulation of Living Systems (2010)Google Scholar
  19. 19.
    Knowles, J.D., Corne, D.W., Fleischer, M.: Bounded archiving using the Lebesgue measure. In: The 2003 Congress on Evolutionary Computation, vol. 4, December 2003Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Tønnes F. Nygaard
    • 1
    Email author
  • Charles P. Martin
    • 1
  • Jim Torresen
    • 1
  • Kyrre Glette
    • 1
  1. 1.University of OsloOsloNorway

Personalised recommendations