A Sonar-based Position Estimation Algorithm for Localization of Mobile Robots

초음파 센서를 이용한 이동로봇의 자기위치 파악 알고리즘

  • Joe, Woong-Yeol (Intelligent System Control Research Center, Korea Institute of Science and Technology (KIST)) ;
  • Oh, Sang-Rok (Intelligent System Control Research Center, Korea Institute of Science and Technology (KIST)) ;
  • Yu, Bum-Jae (Intelligent System Control Research Center, Korea Institute of Science and Technology (KIST)) ;
  • Park, Gwi-Tae (Dept. of Electrical Engineering, Korea University)
  • 조웅열 (한국과학기술연구원 지능제어연구센터) ;
  • 오상록 (한국과학기술연구원 지능제어연구센터) ;
  • 유범재 (한국과학기술연구원 지능제어연구센터) ;
  • 박귀태 (고려대학교)
  • Published : 2002.11.30

Abstract

This paper presents a modified localization scheme of a mobile robot. When it navigates, the position error of a robot is increased and doesn't go to a goal point where the robot intends to go at the beginning. The objective of localization is to estimate the position of a robot precisely. Many algorithms were developed and still are being researched for localization of a mobile robot at present. Among them, a localization algorithm named continuous localization proposed by Schultz has some merits on real-time navigation and is easy to be implemented compared to other localization schemes. Continuous Localization (CL) is based on map-matching algorithm with global and local maps using only ultrasonic sensors for making grid maps. However, CL has some problems in the process of searching the best-scored-map, when it is applied to a mobile robot. We here propose fast and powerful map-matching algorithm for localization of a mobile robot by experiments.

Keywords