Kernel-RLS Aided GPS/INS Integration with Application to Land Vehicle Navigation

In this paper, the authors consider an intelligent method to enhance the performance of a coupled GPS/INS (Global Positioning System/Inertial Navigation System) for land vehicle navigation during GPS outages. Their method is based on the use of Kernel-RLS to intelligently aid the GPS/INS integration system. The Kernel-RLS is a nonlinear version of the recursive least squares (RLS) algorithm. It performs linear regression in a high-dimensional feature space induced by a Mercer kernel and can therefore be used to recursively construct minimum mean-squared-error solutions to nonlinear least-squares problems. The Kernel-RLS can avoid local minimization and over-fitting problems in neural networks. This paper explores the application of the Kernel-RLS to aid the GPS/INS integrated system, especially during GPS outages. To show its applicability, field tests and performance comparison were conducted.

• Corporate Authors:

  ITS Japan

  Tokyo,   Japan 

  ITS America

  1100 17th Street, NW, 12th Floor
  Washington, DC  United States  20036

  ERTICO

  326 Avenue Louis
  Brussels,   Belgium  B-1050
• Authors:
  • Lee, Jaehoon
  • Nam, Dongkyun
  • Kang, Daesung
  • Park, Jooyoung
  • Yoo, Dong Hyun
  • Kim, Do Yoon
• Conference:
  • 17th ITS World Congress
  • Location: Busan , Korea, South
  • Date: 2010-10-25 to 2010-10-29
• Publication Date: 2010

Language

• English

Media Info

• Media Type: Digital/other
• Features: Figures; References; Tables;
• Pagination: 13p
• Monograph Title: 17th ITS World Congress, Busan, 2010: Proceedings

Subject/Index Terms

• TRT Terms: Algorithms; Automobile navigation systems; Global Positioning System; Inertial navigation systems; Least squares method
• Subject Areas: Highways; Operations and Traffic Management; I70: Traffic and Transport;

Filing Info

• Accession Number: 01365995
• Record Type: Publication
• Files: TRIS
• Created Date: Mar 23 2012 8:44AM