| Institut des Systèmes Intelligents et de Robotique |  |
| Institut des Systèmes Intelligents et de Robotique | |
| HAL : hal-00703755, version 1 |
| Fiche détaillée |  Récupérer au format |
 |
| Robotica 2012, Guimaraes : Portugal (2012) |
|
|
|
|
| Towards fast and adaptive optimal control policies for robots: A direct policy search approach |
|
|
Didier Marin 1Olivier Sigaud 1 |
|
|
| (2012) |
|
|
|
| Optimal control methods are generally too expensive to be applied on-line and in real-time to the control of robots. An alternative method consists in tuning a parametrized reactive controller so that it converges to optimal behavior. In this paper we present such a method based on the "direct Policy Search" paradigm to get a cost-efficient control policy for a simulated two degrees-of-freedom planar arm actuated by six muscles. We learn a parametric controller from demonstration using a few near-optimal trajectories. Then we tune the parameters of this controller using two versions of a Cross-Entropy Policy Search method that we compare. Finally, we show that the resulting controller is 20000 times faster than an optimal control method producing the same trajectories. |
|
|
|
|
|
|
|
|
|
|
|
| 1 : | Institut des Systèmes Intelligents et Robotique (ISIR) |
|
|
|
CNRS : UMR7222 – Université Pierre et Marie Curie [UPMC] - Paris VI |
|
|
|
|
|
|
|
|
|
| Domaine | : | Informatique/Intelligence artificielle |
| hal-00703755, version 1 | |
| http://hal.upmc.fr/hal-00703755 | |
| oai:hal.upmc.fr:hal-00703755 | |
| Contributeur : Didier Marin | |
| Soumis le : Lundi 4 Juin 2012, 12:08:46 | |
| Dernière modification le : Lundi 4 Juin 2012, 12:08:46 | |
