Abstract
Bayesian filtering is a general framework for recursively estimating the state of a dynamical system. Key components of each Bayes filter are probabilistic prediction and observation models. This paper shows how non-parametric Gaussian process (GP) regression can be used for learning such models from training data. We also show how Gaussian process models can be integrated into different versions of Bayes filters, namely particle filters and extended and unscented Kalman filters. The resulting GP-BayesFilters can have several advantages over standard (parametric) filters. Most importantly, GP-BayesFilters do not require an accurate, parametric model of the system. Given enough training data, they enable improved tracking accuracy compared to parametric models, and they degrade gracefully with increased model uncertainty. These advantages stem from the fact that GPs consider both the noise in the system and the uncertainty in the model. If an approximate parametric model is available, it can be incorporated into the GP, resulting in further performance improvements. In experiments, we show different properties of GP-BayesFilters using data collected with an autonomous micro-blimp as well as synthetic data.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Abbeel, P., Coates, A., Montemerlo, M., Ng, A., & Thrun, S. (2005). Discriminative training of Kalman filters. In Proceedings of robotics: science and systems (RSS).
Bar-Shalom, Y., Li, X.-R., & Kirubarajan, T. (2001). Estimation with applications to tracking and navigation. New York: Wiley.
Brooks, A., Makarenko, A., & Upcroft, B. (2006). Gaussian process models for sensor-centric robot localisation. In Proceedings of the IEEE international conference on robotics & automation (ICRA) (pp. 56–61).
Csató, L., & Opper, M. (2002). Sparse on-line Gaussian processes. Neural Computation, 14(3), 641–668.
Deshpande, A. D., Ko, J., & Matsuoka, Y. (2009). Anatomically correct testbed hand control: muscle and joint control strategies. In Proceedings of the IEEE international conference on robotics & automation (ICRA).
Doucet, A., de Freitas, J. F. G., Murphy, K., & Russell, S. (2000). Rao-Blackwellised particle filtering for dynamic Bayesian networks. In Proceedings of the conference on uncertainty in artificial intelligence (UAI).
Edakunni, N., Schaal, S., & Vijayakumar, S. (2007). Kernel carpentry for online regression using randomly varying coefficient model. In Proceedings of the international joint conference on artificial intelligence (IJCAI).
Engel, Y., Szabo, P., & Volkinshtein, D. (2006). Learning to control an octopus arm with Gaussian process temporal difference methods. In Y. Weiss, B. Schölkopf, & J. Platt (Eds.), Advances in neural information processing systems 18 (NIPS) (pp. 347–354). Cambridge: MIT Press.
Ferris, B., Hähnel, D., & Fox, D. (2006). Gaussian processes for signal strength-based location estimation. In Proceedings of robotics: science and systems (RSS).
Ferris, B., Fox, D., & Lawrence, N. (2007). WiFi-SLAM using Gaussian process latent variable models. In Proceedings of the international joint conference on artificial intelligence (IJCAI).
Girard, A., Rasmussen, C., Quin̈onero Candela, J., & Murray-Smith, R. (2005). Gaussian process priors with uncertain inputs—application to multiple-step ahead time series forecasting. In Advances in neural information processing systems 15 (NIPS). Cambridge: MIT Press.
Goldberg, P., Williams, C. K. I., & Bishop, C. (1998). Regression with input-dependent noise: A Gaussian process treatment. In Advances in neural information processing systems 10 (NIPS) (pp. 493–499). Cambridge: MIT Press.
Gomes, S. B. V., & Ramos, J. G. Jr. (1998). Airship dynamic modeling for autonomous operation. Proceedings of the IEEE international conference on robotics & automation (ICRA) (p. 4).
Grimes, D., Chalodhorn, R., & Rao, R. (2006). Dynamic imitation in a humanoid robot through nonparametric probabilistic inference. In Proceedings of robotics: science and systems (RSS).
Julier, S., & Uhlmann, J. (1997). A new extension of the Kalman filter to nonlinear systems. In International symposium on aerospace/defense sensing, simulation and controls.
Kersting, K., Plagemann, C., Pfaff, P., & Burgard, W. (2007). Most likely heteroscedastic Gaussian process regression. In Proceedings of the international conference on machine learning (ICML), Corvallis, Oregon, USA, March 2007.
Ko, J., & Fox, D. (2008). GP-BayesFilters: Bayesian filtering using Gaussian process prediction and observation models. In Proceedings of the IEEE/RSJ international conference on intelligent robots and systems (IROS).
Ko, J., & Fox, D. (2009). Learning GP-BayesFilters via Gaussian process latent variable models. In Proceedings of robotics: science and systems (RSS).
Ko, J., Klein, D., Fox, D., & Hähnel, D. (2007a). GP-UKF: unscented Kalman filters with Gaussian process prediction and observation models. In Proceedings of the IEEE/RSJ international conference on intelligent robots and systems (IROS).
Ko, J., Klein, D. J., Fox, D., & Hähnel, D. (2007b). Gaussian processes and reinforcement learning for identification and control of an autonomous blimp. In Proceedings of the IEEE international conference on robotics & automation (ICRA).
Kocijan, J., Girard, A., Banko, B., & Murray-Smith, R. (2003). Dynamic systems identification with Gaussian processes. In Proceedings of 4th IMACS symposium on mathematical modelling (MathMod) (pp. 776–784).
Lawrence, N. (2004). Gaussian process latent variable models for visualisation of high dimensional data. In Advances in neural information processing systems 17 (NIPS) (p. 2004).
Lawrence, N. (2005). Probabilistic non-linear principal component analysis with Gaussian process latent variable models. Journal of Machine Learning Research, 6.
Lawrence, N. D. http://www.dcs.shef.ac.uk/~neil/fgplvm/.
Limketkai, B., Fox, D., & Liao, L. (2007). CRF-filters: discriminative particle filters for sequential state estimation. In Proceedings of the IEEE international conference on robotics & automation (ICRA).
Liu, K., Hertzmann, A., & Popovic, Z. (2005). Learning physics-based motion style with nonlinear inverse optimization. In ACM transactions on graphics (Proceedings of SIGGRAPH).
Nguyen, D., & Peters, J. (2008). Local Gaussian processes regression for real-time model-based robot control. In Proceedings of the IEEE/RSJ international conference on intelligent robots and systems (IROS).
Nott, D. (2006). Semiparametric estimation of mean and variance functions for non-Gaussian data. Computational Statistics, 21(3–4), 603–620.
Paciorek, C. (2003). Nonstationary Gaussian processes for regression and spatial modelling.
Plagemann, C., Fox, D., & Burgard, W. (2007). Efficient failure detection on mobile robots using Gaussian process proposals. In Proceedings of the international joint conference on artificial intelligence (IJCAI).
Plagemann, C., Kersting, K., & Burgard, W. (2008). Nonstationary Gaussian process regression using point estimates of local smoothness. In ECML PKDD ’08: proceedings of the European conference on machine learning and knowledge discovery in databases, part II (pp. 204–219), Antwerp, Belgium. Berlin: Springer. ISBN:978-3-540-87480-5, doi:10.1007/978-3-540-87481-2_14
Quinonero-Candela, J., & Rasmussen, C. E. (2005). A unifying view of sparse approximate Gaussian process regression. Journal of Machine Learning Research, 6, 1939–1959.
Rasmussen, C. E., & Williams, C. K. I. (2005). Gaussian processes for machine learning. Cambridge: MIT Press.
Sanner, R., & Slotine, J. (1991). Stable adaptive control and recursive identification using radial Gaussian networks. In Proceedings of IEEE conference on decision and control (CDC).
Schölkopf, B., Smola, A., Williamson, R., & Bartlett, P. (2000). Letter communicated by John Platt new support vector algorithms.
Seeger, M., & Williams, C. (2003). Fast forward selection to speed up sparse Gaussian process regression. In Workshop on AI and Statistics 9.
Seo, S., Wallat, M., Graepel, T., & Obermayer, K. (2000). Gaussian process regression: Active data selection and test point rejection. In German association for pattern recognition (DAGM) symposium (pp. 27–34).
Smola, A., & Bartlett, P. (2001). Sparse greedy Gaussian process regression. In Advances in neural information processing systems (NIPS) (pp. 619–625). Cambridge: MIT Press.
Snelson, E., & Ghahramani, Z. (2006). Sparse Gaussian processes using pseudo-inputs. In Advances in neural information processing systems (NIPS) (pp. 1257–1264). Cambridge: MIT Press.
Stevens, B. L., & Lewis, F. L. (1992). Aircraft control and simulation. New York: Wiley.
Thrun, S., Burgard, W., & Fox, D. (2005). Probabilistic robotics. Cambridge: MIT Press. ISBN: 0-262-20162-3.
Wang, J., Fleet, D., & Hertzmann, A. (2006). Gaussian process dynamical models. In Advances in neural information processing systems 18 (NIPS) (pp. 1441–1448). Cambridge: MIT Press.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Ko, J., Fox, D. GP-BayesFilters: Bayesian filtering using Gaussian process prediction and observation models. Auton Robot 27, 75–90 (2009). https://doi.org/10.1007/s10514-009-9119-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10514-009-9119-x