Neural Network Optimal Feedback Control with Guaranteed Local Stability

Abstract

Recent research shows that supervised learning can be an effective tool for designing near-optimal feedback controllers for high-dimensional nonlinear dynamic systems. But the behavior of neural network controllers is still not well understood. In particular, some neural networks with high test accuracy can fail to even locally stabilize the dynamic system. To address this challenge we propose several novel neural network architectures, which we show guarantee local asymptotic stability while retaining the approximation capacity to learn the optimal feedback policy semi-globally. The proposed architectures are compared against standard neural network feedback controllers through numerical simulations of two high-dimensional nonlinear optimal control problems: stabilization of an unstable Burgers-type partial differential equation, and altitude and course tracking for an unmanned aerial vehicle. The simulations demonstrate that standard neural networks can fail to stabilize the dynamics even when trained well, while the proposed architectures are always at least locally stabilizing and can achieve near-optimal performance.

arXiv commentarXiv admin note: text overlap with arXiv:2109.07466

References in corpus (14)

• Adam: A Method for Stochastic Optimization
• SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
• DGM: A deep learning algorithm for solving partial differential equations
• Deep Learning Approximation for Stochastic Control Problems
• Adaptive Deep Learning for High-Dimensional Hamilton-Jacobi-Bellman Equations
• Polynomial approximation of high-dimensional Hamilton-Jacobi-Bellman equations and applications to feedback control of semilinear parabolic PDEs
• A Neural Network Approach for High-Dimensional Optimal Control Applied to Multi-Agent Path Finding
• QRnet: optimal regulator design with LQR-augmented neural networks
• Optimal Feedback Law Recovery by Gradient-Augmented Sparse Polynomial Regression
• Aggressive Online Control of a Quadrotor via Deep Network Representations of Optimality Principles
• Algorithms of Data Development For Deep Learning and Feedback Design
• Interplanetary Transfers via Deep Representations of the Optimal Policy and/or of the Value Function
• Neural network optimal feedback control with enhanced closed loop stability
• A Machine Learning Enhanced Algorithm for the Optimal Landing Problem

Cited by in corpus (3)

• Deep neural network approximations for the stable manifolds of the Hamilton-Jacobi-Bellman equations
• YANNs: Y-wise Affine Neural Networks for Exact and Efficient Representations of Piecewise Linear Functions
• Progressive Optimal Path Sampling for Closed-Loop Optimal Control Design with Deep Neural Networks