A new algorithm for the LQR problem with partially unknown dynamics

DOI10.1007/978-3-030-97549-4_37zbMath1487.49039arXiv2105.13723OpenAlexW3164352982MaRDI QIDQ2128456

Agnese Pacifico, Maurizio Falcone, Andrea Pesare

Publication date: 22 April 2022

Full work available at URL: https://arxiv.org/abs/2105.13723

zbMATH Keywords

numerical methodsreinforcement learningGaussian distributionsBayesian regressionLQR problem

Mathematics Subject Classification ID

Linear regression; mixed models (62J05) Probability distributions: general theory (60E05) Learning and adaptive systems in artificial intelligence (68T05) Optimal feedback synthesis (49N35) Linear-quadratic optimal control problems (49N10)

Related Items (1)

Optimal control of ensembles of dynamical systems

Uses Software

• PILCO

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Dual-loop iterative optimal control for the finite horizon LQR problem with unknown dynamics
• A model for system uncertainty in reinforcement learning
• Convergence results for an averaged LQR problem with applications to reinforcement learning
• Convex Optimization: Algorithms and Complexity
• Linear quadratic optimal learning control (LQL)
• Adaptive dynamic programming for finite-horizon optimal control of linear time-varying discrete-time systems
• LQ control of unknown discrete‐time linear systems—A novel approach and a comparison study

This page was built for publication: A new algorithm for the LQR problem with partially unknown dynamics