ImitationFlow: Learning Deep Stable Stochastic Dynamic Systems by Normalizing Flows
Authors:
- 24 October 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 5231-5237
Abstract
We introduce ImitationFlow, a novel Deep generative model that allows learning complex globally stable, stochastic, nonlinear dynamics. Our approach extends the Normalizing Flows framework to learn stable Stochastic Differential Equations. We prove the Lyapunov stability for a class of Stochastic Differential Equations and we propose a learning algorithm to learn them from a set of demonstrated trajectories. Our model extends the set of stable dynamical systems that can be represented by state-of-the-art approaches, eliminates the Gaussian assumption on the demonstrations, and outperforms the previous algorithms in terms of representation accuracy. We show the effectiveness of our method with both standard datasets and a real robot experiment.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- Fast diffeomorphic matching to learn globally asymptotically stable nonlinear dynamical systemsSystems & Control Letters, 2016
- Learning Contracting Nonlinear Dynamics From Human Demonstration for Robot Motion Planning2015
- Learning robot motions with stable dynamical systems under diffeomorphic transformationsRobotics and Autonomous Systems, 2015
- Learning control Lyapunov function to ensure stability of dynamical system-based robot reaching motionsRobotics and Autonomous Systems, 2014
- Neural learning of stable dynamical systems based on data-driven Lyapunov candidates2013
- Learning Stable Nonlinear Dynamical Systems With Gaussian Mixture ModelsIEEE Transactions on Robotics, 2011
- Numerical Solution of Stochastic Differential Equations with Jumps in Finance2010
- Dynamical System Modulation for Robot Learning via Kinesthetic DemonstrationsIEEE Transactions on Robotics, 2008
- Dynamic Movement Primitives -A Framework for Motor Control in Humans and Humanoid Robotics2006
- Locally Weighted Learning1997