Learning to Control an Unstable System with One Minute of Data: Leveraging Gaussian Process Differentiation in Predictive Control

@article{Rodriguez2021LearningTC,
  title={Learning to Control an Unstable System with One Minute of Data: Leveraging Gaussian Process Differentiation in Predictive Control},
  author={Ivan Dario Jimenez Rodriguez and Ugo Rosolia and A. Ames and Yisong Yue},
  journal={2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  year={2021},
  pages={3896-3903}
}
We present a straightforward and efficient way to control unstable robotic systems using an estimated dynamics model. Specifically, we show how to exploit the differentiability of Gaussian Processes to create a state-dependent linearized approximation of the true continuous dynamics that can be integrated with model predictive control. Our approach is compatible with most Gaussian process approaches for system identification, and can learn an accurate model using modest amounts of training data… 
View on IEEE
authors.library.caltech.edu

Figures from this paper

References

SHOWING 1-10 OF 44 REFERENCES
Gaussian Processes for Data-Efficient Learning in Robotics and Control
TLDR
This paper learns a probabilistic, non-parametric Gaussian process transition model of the system and applies it to autonomous learning in real robot and control tasks, achieving an unprecedented speed of learning.
Cautious Model Predictive Control Using Gaussian Process Regression
TLDR
This work describes a principled way of formulating the chance-constrained MPC problem, which takes into account residual uncertainties provided by the GP model to enable cautious control and presents a model predictive control approach that integrates a nominal system with an additive nonlinear part of the dynamics modeled as a GP.
GP-ILQG: Data-driven Robust Optimal Control for Uncertain Nonlinear Dynamical Systems
TLDR
A novel algorithm is proposed that combines data-driven system identification approach (Gaussian Process) with a Differential-Dynamic-Programming-based robust optimal control method (Iterative Linear Quadratic Control) and learns only the difference between the simulator's prediction and actual observations, making it a natural hybrid of simulation and real-world observation.
Learning-Based Model Predictive Control: Toward Safe Learning in Control
TLDR
This research presents a meta-modelling framework that automates the very labor-intensive and therefore time-heavy and expensive process of manually cataloging and cataloging individual neurons to provide real-time information about their levels of activity.
Learning-Based Model Predictive Control for Safe Exploration
TLDR
This paper presents a learning-based model predictive control scheme that can provide provable high-probability safety guarantees and exploits regularity assumptions on the dynamics in terms of a Gaussian process prior to construct provably accurate confidence intervals on predicted trajectories.
Gaussian Process-Based Predictive Control for Periodic Error Correction
TLDR
It is shown that hyperparameter estimation can be performed online using the maximum a posteriori point estimate, which provides an accuracy comparable with sampling methods as soon as enough data to cover the periodic structure has been collected.
Episodic Learning with Control Lyapunov Functions for Uncertain Robotic Systems*
TLDR
A machine learning framework centered around Control Lyapunov Functions to adapt to parametric uncertainty and unmodeled dynamics in general robotic systems and yields a stabilizing quadratic program model-based controller.
Robust Guarantees for Perception-Based Control
TLDR
This work shows that under suitable smoothness assumptions on the perception map and generative model relating state to high-dimensional data, an affine error model is sufficiently rich to capture all possible error profiles, and can be learned via a robust regression problem.
Neural Lander: Stable Drone Landing Control Using Learned Dynamics
TLDR
A novel deep-learning-based robust nonlinear controller (Neural-Lander) that improves control performance of a quadrotor during landing and is the first DNN-based nonlinear feedback controller with stability guarantees that can utilize arbitrarily large neural nets.
Control of Unknown Nonlinear Systems with Linear Time-Varying MPC
TLDR
A non-parametric method is used to estimate the nonlinear dynamics from observed data and this is linearized over time-varying regions of the state space to construct an Affine Time-Varying (ATV) model.
...
...