Welcome to Acta Armamentarii ! Today is Share:

Acta Armamentarii ›› 2020, Vol. 41 ›› Issue (7): 1384-1392.doi: 10.3969/j.issn.1000-1093.2020.07.016

• Paper • Previous Articles     Next Articles

A Multi-beam Sonar-based Graph SLAM Method

DAI Tian, MIAO Lingjuan, SHAO Haijun   

  1. (School of Automation, Beijing Institute of Technology, Beijing 100081, China)
  • Received:2019-07-10 Revised:2019-07-10 Online:2020-09-23

Abstract: Database-referenced navigation systems(DBRNSs) are commonly used in autonomous underwater vehicles as underwater auxiliary navigation systems to correct the cumulative errors of inertial navigation systems. However, the DBRNS cannot run if a high resolution database cannot be obtained. A multi-beam sonar-based graph simultaneous localization and mapping (SLAM) method is proposed. A mismatch detection module is added to the front end to reduce the negative impact caused by the false closed-loop detection results. In the proposed method, the local gray value coding algorithm is used for the closed-loop detection module in order to improve the detection accuracy. Simulations based on the actual seabed terrain base map show that, compared with the general graph SLAM method, the proposed method can achieve higher positioning accuracy. Key

Key words: autonomousunderwatervehicle, simultaneouslocalizationandmapping, multi-beamsonar, inertialnavigationsystem, mismatchdetection

CLC Number: