Geodesic
Reinforcement learning for manipulator control
Russian journal of nonlinear dynamics, Tome 8 (2012) no. 4, pp. 689-704

Voir la notice de l'article provenant de la source Math-Net.Ru

We present method for constructing manipulator control system with reinforcement learning algorithm. We construct learning algorithm which uses information about performed actions and their quality with respect to desired behaviour called «reward». The goal of the learning algorithm is to construct control system maximizing total reward. Learning algorithm and constructed control system were tested on the manipulator collision avoidance problem.
Keywords: reinforcement learning, manipulator, control, Newton–Euler algorithm. 
@article{ND_2012_8_4_a1,
     author = {Nataly P. Koshmanova and Dmitry S. Trifonov and Vladimir E. Pavlovsky},
     title = {Reinforcement learning for manipulator control},
     journal = {Russian journal of nonlinear dynamics},
     pages = {689--704},
     publisher = {mathdoc},
     volume = {8},
     number = {4},
     year = {2012},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/ND_2012_8_4_a1/}
}
TY  - JOUR
AU  - Nataly P. Koshmanova
AU  - Dmitry S. Trifonov
AU  - Vladimir E. Pavlovsky
TI  - Reinforcement learning for manipulator control
JO  - Russian journal of nonlinear dynamics
PY  - 2012
SP  - 689
EP  - 704
VL  - 8
IS  - 4
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/ND_2012_8_4_a1/
LA  - ru
ID  - ND_2012_8_4_a1
ER  - 
%0 Journal Article
%A Nataly P. Koshmanova
%A Dmitry S. Trifonov
%A Vladimir E. Pavlovsky
%T Reinforcement learning for manipulator control
%J Russian journal of nonlinear dynamics
%D 2012
%P 689-704
%V 8
%N 4
%I mathdoc
%U http://geodesic.mathdoc.fr/item/ND_2012_8_4_a1/
%G ru
%F ND_2012_8_4_a1
Nataly P. Koshmanova; Dmitry S. Trifonov; Vladimir E. Pavlovsky. Reinforcement learning for manipulator control. Russian journal of nonlinear dynamics, Tome 8 (2012) no. 4, pp. 689-704. http://geodesic.mathdoc.fr/item/ND_2012_8_4_a1/