Approximate dynamic programming for constrained linear systems: a piecewise quadratic approximation approach

DOI10.1016/j.automatica.2023.111456arXiv2205.10065MaRDI QIDQ6152577

Shengling Shi, Kanghui He, Ton J. J. van den Boom, Bart De Schutter

Publication date: 13 February 2024

Published in: Automatica (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2205.10065

zbMATH Keywords

neural networksreinforcement learningmodel predictive controlvalue function approximationapproximate dynamic programmingconstrained linear quadratic regulation

Mathematics Subject Classification ID

Dynamic programming in optimal control and differential games (49L20) Linear-quadratic optimal control problems (49N10) Model predictive control (93B45)

Cites Work

• Unnamed Item
• Unnamed Item
• On constrained infinite-time linear quadratic optimal control
• Multilayer feedforward networks are universal approximators
• An algorithm for multi-parametric quadratic programming and explicit MPC solutions
• Large scale model predictive control with neural networks and primal active sets
• A predictive safety filter for learning-based control of constrained nonlinear dynamical systems
• Predictive Control for Linear and Hybrid Systems
• Efficient On-Line Computation of Constrained Optimal Control
• Stochastic global optimization methods part II: Multi level methods
• Piecewise convex programs
• The explicit linear quadratic regulator for constrained systems