DocumentCode :
3012754
Title :
Operational Feel Adjustment for Power-Assisted Positioning by an Actor-Critic Method
Author :
Morizono, Tetsuya
Author_Institution :
Dept. of Inf. & Syst. Eng., Fukuoka Inst. of Technol., Fukuoka, Japan
fYear :
2010
fDate :
4-6 Nov. 2010
Firstpage :
743
Lastpage :
748
Abstract :
This paper proposes a method of operational feel adjustment using reinforcement learning for power-assisted positioning. To illustrate difference from the author´s previous study and motivation of the study in this paper, two methods of reinforcement learning, Sarsa and actor-critic methods, are experimentally compared at first with a simple example. Then, operational feel adjustment using an actor-critic method is proposed and examined by experiment. Results obtained in experiment are observed for some discussion, and concluding remark makes mention of some future issues.
Keywords :
learning (artificial intelligence); mobile robots; position control; actor critic method; operational feel adjustment; power assisted positioning; reinforcement learning; Equalizers; Equations; Impedance; Learning; Mathematical model; Service robots; actor-critic; adjustment; operational feel; positioning; power-assist; reinforcement learning; robot;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Broadband, Wireless Computing, Communication and Applications (BWCCA), 2010 International Conference on
Conference_Location :
Fukuoka
Print_ISBN :
978-1-4244-8448-5
Electronic_ISBN :
978-0-7695-4236-2
Type :
conf
DOI :
10.1109/BWCCA.2010.165
Filename :
5631549
Link To Document :
بازگشت