Simplified Q-learning for holonomic mobile robot navigation

Author

Adiprawita, Widyawardana ; Ahmad, Adang Suwandi ; Sembiring, Jaka ; Trilaksono, Bambang R.

Author_Institution

Sch. of Electr. Eng. & Inf., Bandung Inst. of Technol., Bandung, Indonesia

fYear

2011

fDate

8-9 Nov. 2011

Firstpage

64

Lastpage

68

Abstract

In this paper we propose a method of presenting a special case of Value Function as a solution to POMDP in holonomic mobile robot navigation. By using this new method the Value Function complexity will be reduced and more intuitive. The result of this new Value Function is validated with particle filter simulation in Matlab and also experimented physically using a simple autonomous mobile robot built with Lego Mindstorms NXT with 3 ultrasonic sonar and RWTH Mindstorms NXT Toolbox for Matlab to connect the robot to Matlab. This simulation and experiment also incorporate particle filter localization from previous research. The simulation and experiment show that the Value Function can be utilized very well.

Keywords

computational complexity; learning (artificial intelligence); mobile robots; sonar; ultrasonic applications; Lego Mindstorms NXT; Matlab; POMDP; RWTH Mindstorms NXT Toolbox; autonomous mobile robot; holonomic mobile robot navigation; simplified q-learning; ultrasonic sonar; value function complexity; Computer architecture; Educational institutions; Mobile robots; Probabilistic logic; Robot sensing systems; Uncertainty; LEGO Mindstorm NXR; POMDP; Q-Learning; RWTH toolbox; autonomous mobile robot; navigation; value function; value iteration;

fLanguage

English

Publisher

ieee

Conference_Titel

Instrumentation, Communications, Information Technology, and Biomedical Engineering (ICICI-BME), 2011 2nd International Conference on

Conference_Location

Bandung

Print_ISBN

978-1-4577-1167-1

Type

conf

DOI

10.1109/ICICI-BME.2011.6108595

Filename

6108595