Online Learning of Parameterized Uncertain Dynamical Environments With Finite-Sample Guarantees
- 11 November 2020
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems Letters
- Vol. 5 (4) , 1351-1356
Abstract
We present a novel online learning algorithm for a class of unknown and uncertain dynamical environments that are fully observable. First, we obtain a novel probabilistic characterization of systems whose mean behavior is known but which are subject to additive, unknown subGaussian disturbances. This characterization relies on recent concentration of measure results and is given in terms of ambiguity sets. Second, we extend the results to environments whose mean behavior is also unknown but described by a parameterized class of possible mean behaviors. Our algorithm adapts the ambiguity set dynamically by learning the parametric dependence online, and retaining similar probabilistic guarantees with respect to the additive, unknown disturbance. We illustrate the results on a differential-drive robot subject to environmental uncertainty.Keywords
Funding Information
- Office of Naval Research (N00014-19-1-2471)
- Air Force Office of Scientific Research (FA9550-19-1-0235)
This publication has 14 references indexed in Scilit:
- Data-Driven Ambiguity Sets With Probabilistic Guarantees for Dynamic ProcessesIEEE Transactions on Automatic Control, 2020
- Motion Planning Networks: Bridging the Gap Between Learning-Based and Classical Motion PlannersIEEE Transactions on Robotics, 2020
- Non-asymptotic Identification of LTI Systems from a Single Trajectory2019
- Dynamic Evolution of Distributional Ambiguity Sets and Precision Tradeoffs in Data Assimilation2019
- System Identification: A Machine Learning PerspectiveAnnual Review of Control, Robotics, and Autonomous Systems, 2019
- High-Dimensional Probability2018
- On the rate of convergence in Wasserstein distance of the empirical measureProbability Theory and Related Fields, 2014
- Unified Set Membership theory for identification, prediction and filtering of nonlinear systemsAutomatica, 2011
- Learning to Move in Modular Robots using Central Pattern Generators and Online OptimizationThe International Journal of Robotics Research, 2008
- Planning Algorithms2006