Skip to main navigation Skip to search Skip to main content

An Improved Adaptive Monte Carlo Localization (AMCL) for Automated Mobile Robot (AMR)

  • Shan He
  • , Tao Song
  • , Xinkai Wu*
  • *Corresponding author for this work
  • Beihang University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The adaptive Monte Carlo localization (AMCL) algorithm is commonly used for localization tasks for automated mobile robots (AMRs). However, when AMRs move to a feature-less environment, AMCL shows poor performance in localization. We propose an improved AMCL algorithm to improve the accuracy and robustness of the localization for AMR. We first matched the laser scanning points with the pre-built grid map according to the localization result generated from AMCL. Then, we designed a localization credibility estimation (LCE) to evaluate the localization performance, and the match with the higher LCE score was selected and injected into the particle swarm with an adaptive amount to optimize the next estimation process of AMCL. The particle swarm of AMCL gradually converged to the correct location through iterative optimization. Results demonstrate that the proposed AMCL algorithm is superior to the traditional AMCL algorithm and previous improved AMCL in terms of accuracy and robustness.

Original languageEnglish
Title of host publicationCICTP 2022
Subtitle of host publicationIntelligent, Green, and Connected Transportation - Proceedings of the 22nd COTA International Conference of Transportation Professionals
EditorsShanjiang Zhu, Junfeng Jiao, Hongqi Tian, Guangjun Gao, Xiaokun Wang, Yinggui Zhang, Pu Wang, Helai Huang
PublisherAmerican Society of Civil Engineers (ASCE)
Pages171-181
Number of pages11
ISBN (Electronic)9780784484265
DOIs
StatePublished - 2022
Event22nd COTA International Conference of Transportation Professionals, CICTP 2022 - Changsha, Hunan Province, China
Duration: 8 Jul 202211 Jul 2022

Publication series

NameCICTP 2022: Intelligent, Green, and Connected Transportation - Proceedings of the 22nd COTA International Conference of Transportation Professionals

Conference

Conference22nd COTA International Conference of Transportation Professionals, CICTP 2022
Country/TerritoryChina
CityChangsha, Hunan Province
Period8/07/2211/07/22

Fingerprint

Dive into the research topics of 'An Improved Adaptive Monte Carlo Localization (AMCL) for Automated Mobile Robot (AMR)'. Together they form a unique fingerprint.

Cite this