Q-Learning with adaptive state segmentation (QLASS)

Author

Murao, Hajime ; Kitamura, Shinzo

Author_Institution

Dept. of Comput. & Syst. Eng., Kobe Univ., Japan

fYear

1997

fDate

10-11 Jul 1997

Firstpage

179

Lastpage

184

Abstract

Q-learning is an efficient algorithm to acquire adaptive behavior of the robot without a priori knowledge of the sensor space and the task. However, there is a problem in applying the Q-learning to the task in the real world-how to construct the state space suitable for the Q-learning without knowledge of the sensor space? In this paper we propose Q-learning with adaptive state segmentation (QLASS). QLASS provides a method to segment the sensor space incrementally, based on sensor vectors and reinforcement signals. Experimental results show that QLASS can segment the sensor space effectively to accomplish the task. Furthermore, we show the obtained state space reveals the fitness landscape

Keywords

learning (artificial intelligence); mobile robots; path planning; Q-Learning with adaptive state segmentation; adaptive behavior; fitness landscape; reinforcement signals; sensor space; sensor vectors; state space; Humans; Knowledge engineering; Learning; Neural networks; Neurons; Orbital robotics; Robot sensing systems; Sensor systems; State-space methods; Systems engineering and theory;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Intelligence in Robotics and Automation, 1997. CIRA'97., Proceedings., 1997 IEEE International Symposium on

Conference_Location

Monterey, CA

Print_ISBN

0-8186-8138-1

Type

conf

DOI

10.1109/CIRA.1997.613856

Filename

613856