Large Scale Model Predictive Control with Neural Networks and Primal Active Sets

@article{Chen2019LargeSM,
  title={Large Scale Model Predictive Control with Neural Networks and Primal Active Sets},
  author={Steven W. Chen and Tianyu Wang and Nikolay A. Atanasov and Vijay R. Kumar and Manfred Morari},
  journal={Autom.},
  year={2019},
  volume={135},
  pages={109947}
}
[PDF] Semantic Reader
36 Citations
Highly Influential Citations
4
Background Citations
15
Methods Citations
12

Figures from this paper

36 Citations

End-to-End Learning to Warm-Start for Real-Time Quadratic Optimization

,

Approximate Dynamic Programming for Constrained Linear Systems: A Piecewise Quadratic Approximation Approach

This paper introduces an approach combining the two methodologies to overcome their individual limitations, and proposes an ADP method for CLQR problems using Model predictive control and a novel convex and piecewise quadratic neural network.

Learning Models of Model Predictive Controllers using Gradient Data

Design and Simulation of a Machine-learning and Model Predictive Control Approach to Autonomous Race Driving for the F1/10 Platform

An Improved Data Augmentation Scheme for Model Predictive Control Policy Approximation

An improved data augmentation scheme based on predictor-corrector steps that enforces a user-defined level of accuracy, and shows that the error bound of the augmented samples are independent of the size of the neighborhood used for data augmented.

Neural Operators for Bypassing Gain and Control Computations in PDE Backstepping

A framework for eliminating the computation of controller gain functions in PDE control is introduced, and the existence of a DeepONet approximation of the exact nonlinear continuous operator mapping PDE coefficient functions into gain functions is proved.

Model-Free Adaptive Control of Hydrometallurgy Cascade Gold Leaching Process with Input Constraints

Hydrometallurgy technology can directly deal with low grade and complex materials, improve the comprehensive utilization rate of resources, and effectively adapt to the demand of low carbon and…

Deep Neural Network Based Model Predictive Control for Standoff Tracking by a Quadrotor UAV*

The standoff tracking requires an unmanned aerial vehicle (UAV) to loiter in a circular orbit above a target of interest. To achieve it, we propose a deep neural network (DNN) based model predictive…

Guaranteed safe control of systems with parametric uncertainties via neural network controllers

  • B. KargS. Lucia
  • Computer Science
    2022 IEEE 61st Conference on Decision and Control (CDC)
  • 2022
Mixed-integer problems that enable analyzing the behavior of the closed-loop system consisting of the highly nonlinear neural network controller and a linear system with parametric uncertainties are introduced.

Standoff Tracking Using DNN-Based MPC with Implementation on FPGA

The hardware-in-the-loop (HIL) simulation with an FPGA@ 200MHz demonstrates that the DNN-based MPC scheme is a valid alternative to embedded implementations of MPC for addressing complex systems and applications which is impossible for directly solving the MPC optimization problems.

References

SHOWING 1-10 OF 47 REFERENCES

2016) in physics from Haverford College, PA, and the M.S. (2018) in electrical and computer engineering from University of California, San Diego, where he is currently pursuing a Ph.D

  • 2018

Numerical Optimization

Numerical Optimization presents a comprehensive and up-to-date description of the most effective methods in continuous optimization. It responds to the growing interest in optimization in…

Safe and Near-Optimal Policy Learning for Model Predictive Control using Primal-Dual Neural Networks

A novel framework for approximating the explicit MPC law for linear parameter-varying systems using supervised learning that not only learns the control policy, but also a “certificate policy”, that allows us to estimate the sub-optimality of the learned control policy online, during execution-time.

Deep ReLU Networks Have Surprisingly Few Activation Patterns

This work shows empirically that the average number of activation patterns for ReLU networks at initialization is bounded by the total number of neurons raised to the input dimension, and suggests that realizing the full expressivity of deep networks may not be possible in practice, at least with current methods.

On the distribution of points in a cube and the approximate evaluation of integrals

An economic method of computing LPτ-sequences

Random number generation and Quasi-Monte Carlo methods

  • H. Niederreiter
  • Computer Science, Mathematics
    CBMS-NSF regional conference series in applied mathematics
  • 1992
This chapter discusses Monte Carlo methods and Quasi-Monte Carlo methods for optimization, which are used for numerical integration, and their applications in random numbers and pseudorandom numbers.

Introduction to Quasi-Monte Carlo Integration and Applications

Preface.- Notation.- 1 Introduction.- 2 Uniform Distribution Modulo One.- 3 QMC Integration in Reproducing Kernel Hilbert Spaces.- 4 Lattice Point Sets.- 5 (t, m, s)-nets and (t, s)-Sequences.- 6 A…

A deep learning-based approach to robust nonlinear model predictive control

Machine learning-based warm starting of active set methods in embedded model predictive control