Advertisement

Autonomous Robots

, Volume 30, Issue 1, pp 3–23 | Cite as

Learning GP-BayesFilters via Gaussian process latent variable models

  • Jonathan Ko
  • Dieter Fox
Article

Abstract

GP-BayesFilters are a general framework for integrating Gaussian process prediction and observation models into Bayesian filtering techniques, including particle filters and extended and unscented Kalman filters. GP-BayesFilters have been shown to be extremely well suited for systems for which accurate parametric models are difficult to obtain. GP-BayesFilters learn non-parametric models from training data containing sequences of control inputs, observations, and ground truth states. The need for ground truth states limits the applicability of GP-BayesFilters to systems for which the ground truth can be estimated without significant overhead. In this paper we introduce GPBF-Learn, a framework for training GP-BayesFilters without ground truth states. Our approach extends Gaussian Process Latent Variable Models to the setting of dynamical robotics systems. We show how weak labels for the ground truth states can be incorporated into the GPBF-Learn framework. The approach is evaluated using a difficult tracking task, namely tracking a slotcar based on inertial measurement unit (IMU) observations only. We also show some special features enabled by this framework, including time alignment, and control replay for both the slotcar, and a robotic arm.

Keywords

Gaussian process System identification Bayesian filtering Time alignment System control Machine learning 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Abbeel, P., Dolgov, D., Ng, A., & Thrun, S. (2008). Apprenticeship learning for motion planning with application to parking lot navigation. In Proc. of the IEEE/RSJ international conference on intelligent robots and systems, IROS. Google Scholar
  2. Boots, B., Siddiqi, S., & Gordon, G. (2009). Closing the learning-planning loop with predictive state representations. http://arxiv.org/abs/0912.2385
  3. Bowling, M., Wilkinson, D., Ghodsi, A., & Milstein, A. (2005). Subjective localization with action respecting embedding. In Proc. of the international symposium of robotics research, ISRR. Google Scholar
  4. Coates, A., Abbeel, P., & Ng, A. (2008). Learning for control from multiple demonstrations. In Proc. of the international conference on machine learning, ICML. Google Scholar
  5. Deisenroth, M., Huber, M., & Hanebeck, U. (2009). Analytic moment-based Gaussian process filtering. In Proc. of the international conference on machine learning, ICML (pp. 225–232). New York: ACM. Google Scholar
  6. Ekvall, S., & Kragic, D. (2004). Interactive grasp learning based on human demonstration. In Proc. of the IEEE international conference on robotics & automation, ICRA (pp. 3519–3524). CrossRefGoogle Scholar
  7. Engel, Y., Szabo, P., & Volkinshtein, D. (2006). Learning to control an octopus arm with Gaussian process temporal difference methods. In Advances in neural information processing systems, NIPS (Vol. 18). Google Scholar
  8. Ferris, B., Hähnel, D., & Fox, D. (2006). Gaussian processes for signal strength-based location estimation. In Proc. of robotics: science and systems, RSS. Google Scholar
  9. Ferris, B., Fox, D., & Lawrence, N. (2007). WiFi-SLAM using Gaussian process latent variable models. In Proc. of the international joint conference on artificial intelligence, IJCAI. Google Scholar
  10. Grimes, D., & Rao, R. (2008). Learning nonparametric policies by imitation. In Proc. of the IEEE/RSJ international conference on intelligent robots and systems, IROS (pp. 2022–2028). Google Scholar
  11. Hsu, E., Pulli, K., & Popović, J. (2005). Style translation for human motion. ACM Transactions on Graphics, 24, 1082–1089. CrossRefGoogle Scholar
  12. Hsu, E., da Silva, M., & Popovic, J. (2007). Guided time warping for motion editing. In Symposium on computer animation ’07 proceedings (pp. 45–52). Aire-la-Ville: Eurographics Association. Google Scholar
  13. Kawahara, Y., Yairi, T., & Machida, K. (2007). A kernel subspace method by stochastic realization for learning nonlinear dynamical systems. In B. Schölkopf, J. Platt, & T. Hoffman (Eds.), Advances in neural information processing systems (Vol. 19, pp. 665–672). Cambridge: MIT Press. Google Scholar
  14. Kersting, K., Plagemann, C., Pfaff, P., & Burgard, W. (2007). Most likely heteroscedastic Gaussian process regression. In Proc. of the international conference on machine learning, ICML. Google Scholar
  15. Ko, J., & Fox, D. (2008). GP-BayesFilters: Bayesian filtering using Gaussian process prediction and observation models. In Proc. of the IEEE/RSJ international conference on intelligent robots and systems, IROS. Google Scholar
  16. Ko, J., & Fox, D. (2009). Learning GP-Bayesfilters via Gaussian process latent variable models. In Proc. of robotics: science and systems, RSS. Google Scholar
  17. Ko, J., Klein, D., Fox, D., & Hähnel, D. (2007). Gaussian processes and reinforcement learning for identification and control of an autonomous blimp. In Proc. of the IEEE international conference on robotics & automation, ICRA. Google Scholar
  18. Ko, J., Klein, D., Fox, D., & Hähnel, D. (2007). GP-UKF: Unscented Kalman filters with Gaussian process prediction and observation models. In Proc. of the IEEE/RSJ international conference on intelligent robots and systems, IROS. Google Scholar
  19. Lawrence, N. (2003). Gaussian process latent variable models for visualization of high dimensional data. In Advances in neural information processing systems, NIPS. Google Scholar
  20. Lawrence, N. (2005). Probabilistic non-linear principal component analysis with Gaussian process latent variable models. Journal of Machine Learning Research, 6, 1783–1816. MathSciNetGoogle Scholar
  21. Lawrence, N., & Moore, A. J. (2007). Hierarchical Gaussian process latent variable models. In Proc. of the international conference on machine learning, ICML. Google Scholar
  22. Lawrence, N., & Quiñonero Candela, J. (2006). Local distance preservation in the GP-LVM through back constraints. In Proc. of the international conference on machine learning, ICML. Google Scholar
  23. Littman, M., Sutton, R., & Singh, S. (2001). Predictive representations of state. In Advances in neural information processing systems, NIPS (Vol. 14, pp. 1555–1561). Cambridge: MIT Press. Google Scholar
  24. Ljung, L. (1987). System identification. New York: Prentice Hall. zbMATHGoogle Scholar
  25. Meyer, C. D. (Ed.) (2000). Matrix analysis and applied linear algebra. Philadelphia: Society for Industrial and Applied Mathematics. zbMATHGoogle Scholar
  26. Nguyen-Tuong, D., Seeger, M., & Peters, J. (2008). Local Gaussian process regression for real time online model learning and control. In Advances in neural information processing systems, NIPS (Vol. 22). Google Scholar
  27. Plagemann, C., Fox, D., & Burgard, W. (2007). Efficient failure detection on mobile robots using Gaussian process proposals. In Proc. of the international joint conference on artificial intelligence, IJCAI. Google Scholar
  28. Rabiner, L., Rosenberg, A., & Levinson, S. (1978). Considerations in dynamic time warping algorithms for discrete word recognition. IEEE Transactions on Acoustics, Speech and Signal Processing, 26(6), 575–582. zbMATHCrossRefGoogle Scholar
  29. Rahimi, A., & Recht, B. (2007). Random features for large-scale kernel machines. In Advances in neural information processing systems, NIPS. Google Scholar
  30. Rasmussen, C. E., & Williams, C. K. I. (2005). Gaussian processes for machine learning. Cambridge: MIT Press. Google Scholar
  31. Schmill, M., Oates, T., & Cohen, P. (1999). Learned models for continuous planning. In Proceedings of uncertainty 99: the 7th international workshop on artificial intelligence and statistics (pp. 278–282). Los Altos: Kaufmann. Google Scholar
  32. Sjöberg, J., Zhang, Q., Ljung, L., Benveniste, A., Deylon, B., Glorennec, P., Hjalmarsson, H., & Juditsky, A. (1995). Nonlinear black-box modeling in system identification: a unified overview. Automatica, 31, 1691–1724. zbMATHCrossRefGoogle Scholar
  33. Snelson, E., & Ghahramani, Z. (2006). Sparse Gaussian processes using pseudo-inputs. In Advances in neural information processing systems, NIPS (Vol. 18). Google Scholar
  34. Thrun, S., Burgard, W., & Fox, D. (2005). Probabilistic robotics. Cambridge: MIT Press. ISBN 0-262-20162-3. zbMATHGoogle Scholar
  35. Urtasun, R., Fleet, D., & Fua, P. (2006). Gaussian process dynamical models for 3D people tracking. In Proc. of the IEEE computer society conference on computer vision and pattern recognition, CVPR. Google Scholar
  36. Van Overschee, P., & De Moor, B. (1996). Subspace identification for linear systems: theory, implementation, applications. Norwell: Kluwer Academic. zbMATHGoogle Scholar
  37. Verdult, V., Suykens, J., Boets, J., Goethals, I., De Moor, B., & Leuven, K. (2004). Least squares support vector machines for kernel in nonlinear state-space identification. In Proceedings of the 16th international symposium on mathematical theory of networks and systems, MTNS2004. Google Scholar
  38. Wang, J., Fleet, D., & Hertzmann, A. (2008). Gaussian process dynamical models for human motion. In IEEE transactions on pattern analysis and machine intelligence, PAMI. Google Scholar
  39. Zhou, F., & De la Torre, F. (2009). Canonical time warping for alignment of human behavior. In Advances in neural information processing systems, NIPS. Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2010

Authors and Affiliations

  1. 1.Department of Computer Science & EngineeringUniversity of WashingtonSeattleUSA
  2. 2.Intel Labs SeattleIntel Corp.SeattleUSA

Personalised recommendations