MAM5/M2-INUM

Commande optimale

2024-25

Discussion

Plan du cours

1. Introduction
2. Méthodes directes
3. Principe du maximum de Pontrjagin
4. Méthode de tir
5. Model predictive control
6. Reinforcement learning

CM 1

CM 2

CM 3

CM 4

CM 5

CM 6

CM 7

TP 1 - Navigation (solution) (solution 3 arcs)

TP 2 - Tir simple

TP 3 - MPC

TP 4 - Hexapawn

TP 5 - OpenAI gym: taxi!

Organisation et intervenants

• 10 séances de 3H
• J.-B. Caillau

Évaluation

• 1 EX partiel (coeff. 1)
• 1 EX terminal (coeff. 1)

Exam CC no. 1 (2023-24)

Exam CC no. 2 (2023-24)

Références

Bertsekas, D. A course in reinforcement learning. Athena scientific, 2023

Borkar, X.; Meyn, S. P. The ODE method for convergence of stochastic approximation and reinforcement learning. SIAM J. Control Optim. 38 (2000), no. 2, 447-469

Ekeland I. Mathematics in the social sciences. Torino lectures, 2022

Evans, L. C. An introduction to mathematical optimal control theory. Univ. California, 2008

Fleming, W. H. ; Rishel, R. W. Deterministic and stochastic optimal control. Springer, 1975

Gardner, M. The unexpected hanging and other mathematical diversions. University of Chicago Press, 1991

Recht, B. A tour of reinforcement learning : the view from continuous control. arXiv :1806.09460, 2018

Sutton, R. S. ; Barto, A. G. Reinforcement learning : an introduction, MIT press, 2015

Trélat, E. Contrôle optimal, théorie et applications. Vuibert, 2005