Learning Deep Dynamical Systems using Stable Neural ODEs (2404.10622v2)
Abstract: Learning complex trajectories from demonstrations in robotic tasks has been effectively addressed through the utilization of Dynamical Systems (DS). State-of-the-art DS learning methods ensure stability of the generated trajectories; however, they have three shortcomings: a) the DS is assumed to have a single attractor, which limits the diversity of tasks it can achieve, b) state derivative information is assumed to be available in the learning process and c) the state of the DS is assumed to be measurable at inference time. We propose a class of provably stable latent DS with possibly multiple attractors, that inherit the training methods of Neural Ordinary Differential Equations, thus, dropping the dependency on state derivative information. A diffeomorphic mapping for the output and a loss that captures time-invariant trajectory similarity are proposed. We validate the efficacy of our approach through experiments conducted on a public dataset of handwritten shapes and within a simulated object manipulation task.
- Sylvain Calinon “Learning from demonstration (programming by demonstration)” In Encyclopedia of robotics Springer Berlin/Heidelberg, Germany, 2018, pp. 1–8
- “Collaborative interaction models for optimized human-robot teamwork” In 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020, pp. 11221–11228 IEEE
- “Human-in-the-Loop Optimization of Active Back-Support Exoskeleton Assistance Via Lumbosacral Joint Torque Estimation” In 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2023, pp. 6090–6096 IEEE
- Stefan Schaal “Is imitation learning the route to humanoid robots?” In Trends in cognitive sciences 3.6 Elsevier, 1999, pp. 233–242
- “Mixture of attractors: A novel movement primitive representation for learning motor skills from demonstrations” In IEEE Robotics and Automation Letters 3.2 IEEE, 2018, pp. 926–933
- S Mohammad Khansari-Zadeh and Aude Billard “Learning stable nonlinear dynamical systems with gaussian mixture models” In IEEE Transactions on Robotics 27.5 IEEE, 2011, pp. 943–957
- S Mohammad Khansari-Zadeh and Aude Billard “Learning control Lyapunov function to ensure stability of dynamical system-based robot reaching motions” In Robotics and Autonomous Systems 62.6 Elsevier, 2014, pp. 752–765
- “Euclideanizing flows: Diffeomorphic reduction for learning stable dynamical systems” In Learning for Dynamics and Control, 2020, pp. 630–639 PMLR
- “Dynamical movement primitives: learning attractor models for motor behaviors” In Neural computation 25.2 MIT Press One Rogers Street, Cambridge, MA 02142-1209, USA journals-info …, 2013, pp. 328–373
- J Zico Kolter and Gaurav Manek “Learning stable deep dynamics models” In Advances in neural information processing systems 32, 2019
- Klaus Neumann and Jochen J Steil “Learning robot motions with stable dynamical systems under diffeomorphic transformations” In Robotics and Autonomous Systems 70 Elsevier, 2015, pp. 1–15
- Jiechao Zhang, Hadi Beik Mohammadi and Leonel Rozo “Learning Riemannian Stable Dynamical Systems via Diffeomorphisms” In 6th Annual Conference on Robot Learning, 2022
- “Neural ordinary differential equations” In Advances in neural information processing systems 31, 2018
- “Learning Robust State Observers using Neural ODEs” In Learning for Dynamics and Control Conference, 2023, pp. 208–219 PMLR
- Lucas Böttcher, Nino Antulov-Fantulin and Thomas Asikis “AI Pontryagin or how artificial neural networks learn to control dynamical systems” In Nature communications 13.1 Nature Publishing Group UK London, 2022, pp. 333
- “Near-optimal control of dynamical systems with neural ordinary differential equations” In Machine Learning: Science and Technology 3.4 IOP Publishing, 2022, pp. 045004
- “Deeppose: Human pose estimation via deep neural networks” In Proceedings of the IEEE conference on computer vision and pattern recognition, 2014, pp. 1653–1660
- Ivan Dario Jimenez Rodriguez, Aaron Ames and Yisong Yue “LyaNet: A Lyapunov framework for training neural ODEs” In International Conference on Machine Learning, 2022, pp. 18687–18703 PMLR
- Spencer M Richards, Felix Berkenkamp and Andreas Krause “The lyapunov neural network: Adaptive stability certification for safe learning of dynamical systems” In Conference on Robot Learning, 2018, pp. 466–476 PMLR
- Laurent Dinh, Jascha Sohl-Dickstein and Samy Bengio “Density estimation using real nvp” In arXiv preprint arXiv:1605.08803, 2016
- “Using the averaged Hausdorff distance as a performance measure in evolutionary multiobjective optimization” In IEEE Transactions on Evolutionary Computation 16.4 IEEE, 2012, pp. 504–522
- Patrick Kidger “On neural differential equations” In arXiv preprint arXiv:2202.02435, 2022
- “robosuite: A Modular Simulation Framework and Benchmark for Robot Learning” In arXiv preprint arXiv:2009.12293, 2020
- “What Matters in Learning from Offline Human Demonstrations for Robot Manipulation” In arXiv preprint arXiv:2108.03298, 2021
- “Roboturk: A crowdsourcing platform for robotic skill learning through imitation” In Conference on Robot Learning, 2018, pp. 879–893 PMLR
- Bibit Bianchini, Mathew Halm and Michael Posa “Simultaneous Learning of Contact and Continuous Dynamics” In Conference on Robot Learning, 2023, pp. 3966–3978 PMLR
- Daniel E Koditschek and Elon Rimon “Robot navigation functions on manifolds with boundary” In Advances in applied mathematics 11.4 Elsevier, 1990, pp. 412–442
- Jorge Cortes “Discontinuous dynamical systems” In IEEE Control Systems Magazine 28.3, 2008, pp. 36–73 DOI: 10.1109/MCS.2008.919306