DocumentCode :
596455
Title :
Space recognition in multi-floor building environment
Author :
Park, Soojin ; Sungbae Cho ; Youngouk Kim
Author_Institution :
Sch. of Comput. Sci. Eng., Yonsei Univ., Seoul, South Korea
fYear :
2012
fDate :
26-28 Nov. 2012
Firstpage :
439
Lastpage :
441
Abstract :
Space recognition is one of the essential techniques for service robot. The purpose of this paper is finding out where the floor is collect in similar structural multi-floor building environment. We propose space recognition which is dividing spaces to similar space group and distinct space group and give different weight to each group and calculate state probability. Because our algorithm contains pre-state probability, it is robust for temporary recognition error. In order to evaluate our proposed method, we compared a method which is not applied weighted space recognition and a method which is applied weighted space method. The experiment result shows that our method is suitable for service robot in multi-floor building environment.
Keywords :
probability; robot vision; service robots; distinct space group; prestate probability; service robot; structural multifloor building environment; temporary recognition error; weighted space recognition; Ambient intelligence; Robots; Multi-floor environment; Space recognition;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Ubiquitous Robots and Ambient Intelligence (URAI), 2012 9th International Conference on
Conference_Location :
Daejeon
Print_ISBN :
978-1-4673-3111-1
Electronic_ISBN :
978-1-4673-3110-4
Type :
conf
DOI :
10.1109/URAI.2012.6463034
Filename :
6463034
Link To Document :
بازگشت