Abstract
Designing gaits and corresponding control policies is a key challenge in robot locomotion. Even with a viable controller parametrization, finding near-optimal parameters can be daunting. Typically, this kind of parameter optimization requires specific expert knowledge and extensive robot experiments. Automatic black-box gait optimization methods greatly reduce the need for human expertise and time-consuming design processes. Many different approaches for automatic gait optimization have been suggested to date. However, no extensive comparison among them has yet been performed. In this article, we thoroughly discuss multiple automatic optimization methods in the context of gait optimization. We extensively evaluate Bayesian optimization, a model-based approach to black-box optimization under uncertainty, on both simulated problems and real robots. This evaluation demonstrates that Bayesian optimization is particularly suited for robotic applications, where it is crucial to find a good set of gait parameters in a small number of experiments.
Similar content being viewed by others
References
Chernova, S., Veloso, M.: An evolutionary approach to gait learning for four-legged robots. In: Intelligent Robots and Systems (IROS), vol. 3, pp. 2562–2567. IEEE (2004)
Gibbons, P., Mason, M., Vicente, A., Bugmann, G., Culverhouse, P.: Optimisation of dynamic gait for small bipedal robots. In: Proc. 4th Workshop on Humanoid Soccer Robots (Humanoids 2009), pp. 9–14 (2009)
Kulk, J., Welsh, J.: Evaluation of walk optimisation techniques for the NAO robot. In: Humanoids 2011, pp. 306–311 (2011)
Kushner, H.J.: A new method of locating the maximum point of an arbitrary multipeak curve in the presence of noise. J. Basic Eng. 86, 97 (1964)
Jones, D.R.: A taxonomy of global optimization methods based on response surfaces. J. Global Optim. 21, 345–383 (2001)
Osborne, M.A., Garnett, R., Roberts, S.J.: Gaussian processes for global optimization. In: Learning and Intelligent Optimization (LION), pp. 1–15 (2009)
Brochu, E., Cora, V.M., De Freitas, N.: A tutorial on Bayesian optimization of expensive cost functions, with application to active user modeling and hierarchical reinforcement learning. arXiv preprint arXiv:1012.2599 (2010)
Garnett, R., Osborne, M.A., Roberts, S.J.: Bayesian optimization for sensor set selection. In: Proceedings of the 9th ACM/IEEE International Conference on Information Processing in Sensor Networks. IPSN ’10, pp 209–219. ACM, New York (2010)
Lizotte, D.J., Wang, T., Bowling, M., Schuurmans, D.: Automatic gait optimization with Gaussian process regression. In: International Joint Conference on Artificial Intelligence (IJCAI), pp. 944–949 (2007)
Tesch, M., Schneider, J., Choset, H.: Using response surfaces and expected improvement to optimize snake robot gait parameters. In: International Conference on Intelligent Robots and Systems, pp 1069–1074. IEEE, IROS (2011)
Hutter, F., Hoos, H.H., Leyton-Brown, K.: Sequential model-based optimization for general algorithm configuration. In: Learning and Intelligent Optimization (LION), pp. 507–523. Springer (2011)
Snoek, J., Larochelle, H., Adams, R.P.: Practical bayesian optimization of machine learning algorithms. In: Advances in Neural Information Processing Systems (NIPS) (2012)
Calandra, R., Seyfarth, A., Peters, J., Deisenroth, M.P.: An experimental comparison of Bayesian optimization for bipedal locomotion. In: International Conference on Robotics and Automation (ICRA) (2014)
Calandra, R., Gopalan, N., Seyfarth, A., Peters, J., Deisenroth, M.P.: Bayesian gait optimization for bipedal locomotion. In: Learning and Intelligent Optimization (LION), pp. 274–290 (2014)
Brooks, S.H.: A discussion of random methods for seeking maxima. Oper. Res. 6, 244–251 (1958)
Bergstra, J., Bengio, Y.: Random search for hyper-parameter optimization. J. Mach. Learn. Res. (JMLR) 13, 281–305 (2012)
Yamane, K.: Geometry and biomechanics for locomotion synthesis and control. In: Modeling, Simulation and Optimization of Bipedal Walking. Volume 18 of Cognitive Systems Monographs, pp. 273–287. Springer (2013)
Tedrake, R., Zhang, T., Seung, H.: Stochastic policy gradient reinforcement learning on a simple 3D biped. In: International Conference on Intelligent Robots and Systems (IROS), pp. 2849–2854 (2004)
Tang, Z., Zhou, C., Sun, Z.: Humanoid walking gait optimization using ga-based neural network. In: Advances in Natural Computation. Volume 3611 of Lecture Notes in Computer Science, pp. 252–261 . Springer (2005)
Niehaus, C., Röfer, T., Laue, T.: Gait optimization on a humanoid robot using particle swarm optimization. In: Proceedings of the Second Workshop on Humanoid Soccer Robots in conjunction with the (2007)
Hemker, T., Stelzer, M., von Stryk, O., Sakamoto, H.: Efficient walking speed optimization of a humanoid robot. Int. J. Robot. Res. (IJRR) 28, 303–314 (2009)
Geng, T., Porr, B., Wörgötter, F.: Fast biped walking with a sensor-driven neuronal controller and real-time online learning. Int. J. Robot. Res. (IJRR) 25, 243–259 (2006)
Lizotte, D.J., Greiner, R., Schuurmans, D.: An experimental methodology for response surface optimization methods. J. Global Optim. 53, 699–736 (2012)
Jones, D.R., Schonlau, M., Welch, W.J.: Efficient global optimization of expensive black-box functions. J. Global Optim. 13, 455–492 (1998)
Huang, D., Allen, T.T., Notz, W.I., Zeng, N.: Global optimization of stochastic black-box systems via sequential kriging meta-models. J. Global Optim. 34, 441–466 (2006)
Rasmussen, C.E., Williams, C.K.I.: Gaussian Processes for Machine Learning. The MIT Press (2006)
Mockus, J., Tiesis, V., Zilinskas, A.: The application of Bayesian methods for seeking the extremum. Towards Global Optim 2, 117–129 (1978)
Cox, D.D., John, S.: SDO: A statistical method for global optimization. Multidisciplinary Design Optimization: State of the Art, pp. 315–329 (1997)
Hennig, P., Schuler, C.J.: Entropy search for information-efficient global optimization. J. Mach. Learn. Res. (JMLR) 13, 1809–1837 (2012)
Srinivas, N., Krause, A., Kakade, S., Seeger, M.: Gaussian process optimization in the bandit setting: No regret and experimental design. In: International Conference on Machine Learning (ICML), Omnipress, pp. 1015–1022 (2010)
Jones, D.R., Perttunen, C.D., Stuckman, B.E.: Lipschitzian optimization without the Lipschitz constant. J. Optim. Theory Appl 79, 157–181 (1993)
Byrd, R.H., Lu, P., Nocedal, J., Zhu, C.: A limited memory algorithm for bound constrained optimization. SIAM J. Sci. Comput 16, 1190–1208 (1995)
Hansen, N., Ostermeier, A.: Completely derandomized self-adaptation in evolution strategies. Evol. Comput 9, 159–195 (2001)
Bertsekas, D.P.: Dynamic Programming and Optimal Control. 3rd edn. Athena Scientific (2007)
Renjewski, D.: An engineering contribution to human gait biomechanics. PhD thesis, TU Ilmenau (2012)
Renjewski, D., Seyfarth, A.: Robots in human biomechanics - a study on ankle push-off in walking. Bioinspiration Biomimetics 7, 036005 (2012)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Calandra, R., Seyfarth, A., Peters, J. et al. Bayesian optimization for learning gaits under uncertainty. Ann Math Artif Intell 76, 5–23 (2016). https://doi.org/10.1007/s10472-015-9463-9
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10472-015-9463-9