Deep Lagrangian Networks: Using Physics as Model Prior for Deep Learning

Deep learning has achieved astonishing results on many tasks with large amounts of data and generalization within the proximity of training data. For many important real-world applications, these requirements are unfeasible and additional prior knowledge on the task domain is required to overcome the resulting problems. In particular, learning physics models for model-based control requires robust extrapolation from fewer samples - often collected online in real-time - and model errors may lead to drastic damages of the system. Directly incorporating physical insight has enabled us to obtain a novel deep model learning approach that extrapolates well while requiring fewer samples. As a first example, we propose Deep Lagrangian Networks (DeLaN) as a deep network structure upon which Lagrangian Mechanics have been imposed. DeLaN can learn the equations of motion of a mechanical system (i.e., system dynamics) with a deep network efficiently while ensuring physical plausibility. The resulting DeLaN network performs very well at robot tracking control. The proposed method did not only outperform previous model learning approaches at learning speed but exhibits substantially improved and more robust extrapolation to novel trajectories and learns online in real-time

[1]  H. Hemami,et al.  An approach to analyzing biped locomotion dynamics and designing robot locomotion controls , 1977 .

[2]  H. Hemami,et al.  Modeling and control of constrained dynamic systems with application to biped locomotion in the frontal plane , 1979 .

[3]  J. Y. S. Luh,et al.  On-Line Computational Scheme for Mechanical Manipulators , 1980 .

[4]  W. Book Recursive Lagrangian Dynamics of Flexible Manipulator Arms , 1984 .

[5]  Christopher G. Atkeson,et al.  Estimation of Inertial Parameters of Manipulator Loads and Links , 1986 .

[6]  M. Spong Modeling and Control of Elastic Joint Robots , 1987 .

[7]  Benjamin F. Hobbs,et al.  Is optimization optimistically biased , 1989 .

[8]  Leonard S. Haynes,et al.  On the dynamic model and kinematic analysis of a class of Stewart platforms , 1992, Robotics Auton. Syst..

[9]  Reymond Clavel,et al.  The Lagrange-based model of Delta-4 robot dynamics , 1992, Robotersysteme.

[10]  Frank L. Lewis,et al.  The singularities and dynamics of a Stewart platform manipulator , 1993, J. Intell. Robotic Syst..

[11]  M. Jansen Learning an Accurate Neural Model of the Dynamics of a Typical Industrial Robot , 1994 .

[12]  Carlos Canudas de Wit,et al.  Theory of Robot Control , 1996 .

[13]  J. Doyle,et al.  Robust and optimal control , 1995, Proceedings of 35th IEEE Conference on Decision and Control.

[14]  Dimitrios I. Fotiadis,et al.  Artificial neural networks for solving ordinary and partial differential equations , 1997, IEEE Trans. Neural Networks.

[15]  Dimitris G. Papageorgiou,et al.  Neural-network methods for boundary value problems with irregular boundaries , 2000, IEEE Trans. Neural Networks Learn. Syst..

[16]  Mitsuo Kawato,et al.  MOSAIC Model for Sensorimotor Learning and Control , 2001, Neural Computation.

[17]  A. Albu-Schäffer Regelung von Robotern mit elastischen Gelenken am Beispiel der DLR-Leichtbauarme , 2002 .

[18]  Marko Bacic,et al.  Model predictive control , 2003 .

[19]  Stefan Schaal,et al.  Scalable Techniques from Nonparametric Statistics for Real Time Robot Learning , 2002, Applied Intelligence.

[20]  J. Kocijan,et al.  Gaussian process model based predictive control , 2004, Proceedings of the 2004 American Control Conference.

[21]  Jun Nakanishi,et al.  A Bayesian Approach to Nonlinear Parameter Identification for Rigid Body Dynamics , 2006, Robotics: Science and Systems.

[22]  Nicolas Schweighofer,et al.  Local Online Support Vector Regression for Learning Control , 2007, 2007 International Symposium on Computational Intelligence in Robotics and Automation.

[23]  Marc Toussaint,et al.  Modelling motion primitives and their timing in biologically executed movements , 2007, NIPS.

[24]  Roy Featherstone,et al.  Rigid Body Dynamics Algorithms , 2007 .

[25]  J.P. Ferreira,et al.  Simulation control of a biped robot with Support Vector Regression , 2007, 2007 IEEE International Symposium on Intelligent Signal Processing.

[26]  Jan Peters,et al.  Model Learning with Local Gaussian Process Regression , 2009, Adv. Robotics.

[27]  Darwin G. Caldwell,et al.  Learning and Reproduction of Gestures by Imitation , 2010, IEEE Robotics & Automation Magazine.

[28]  Jan Peters,et al.  Using model knowledge for learning inverse dynamics , 2010, 2010 IEEE International Conference on Robotics and Automation.

[29]  Aude Billard,et al.  Learning Stable Nonlinear Dynamical Systems With Gaussian Mixture Models , 2011, IEEE Transactions on Robotics.

[30]  Jan Peters,et al.  Model learning for robot control: a survey , 2011, Cognitive Processing.

[31]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[32]  Carlos Bordons Alba,et al.  Model Predictive Control , 2012 .

[33]  Shane Legg,et al.  Human-level control through deep reinforcement learning , 2015, Nature.

[34]  Ross A. Knepper,et al.  DeepMPC: Learning Deep Latent Features for Model Predictive Control , 2015, Robotics: Science and Systems.

[35]  Paris Perdikaris,et al.  Physics Informed Deep Learning (Part II): Data-driven Discovery of Nonlinear Partial Differential Equations , 2017, ArXiv.

[36]  Sami Haddadin,et al.  First-order-principles-based constructive network topologies: An application to robot inverse dynamics , 2017, 2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids).

[37]  Demis Hassabis,et al.  Mastering the game of Go without human knowledge , 2017, Nature.

[38]  Jan Peters,et al.  Learning inverse dynamics models in O(n) time with LSTM networks , 2017, 2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids).

[39]  Paris Perdikaris,et al.  Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations , 2017, ArXiv.

[40]  George E. Karniadakis,et al.  Hidden physics models: Machine learning of nonlinear partial differential equations , 2017, J. Comput. Phys..

[41]  Raia Hadsell,et al.  Graph networks as learnable physics engines for inference and control , 2018, ICML.

[42]  Bin Dong,et al.  PDE-Net: Learning PDEs from Data , 2017, ICML.

[43]  Justin A. Sirignano,et al.  DGM: A deep learning algorithm for solving partial differential equations , 2017, J. Comput. Phys..