State inequality constraint based method for rail navigation

Author

Acharya, Arunasish ; Sadhu, Smita ; Ghoshal, T.K.

Author_Institution

Sch. of Educ. Technol., Jadavpur Univ., Kolkata, India

fYear

2010

fDate

17-19 Dec. 2010

Firstpage

1

Lastpage

6

Abstract

This paper deals with a rail navigation problem in GPS dark regions where only accelerometer and odometer data are available. A state inequality constraint based method for estimation with a Bayesian framework and Box-Gaussian approximation has been proposed. The method makes use of sequential pseudo-measurement update. A variant of the method with rejection of outlier measurements is also proposed. These methods are compared with the unconstrained estimation method as well as Gaussian pseudo-measurement method with measurement outlier rejection. The comparative study has been done using Monte Carlo simulations.

Keywords

Bayes methods; Gaussian processes; Global Positioning System; Monte Carlo methods; railway engineering; Bayesian framework; GPS dark regions; Monte Carlo simulations; accelerometer data; box-Gaussian approximation; odometer data; rail navigation; state inequality constraint; Accelerometers; Estimation; Filtering; Navigation; Noise; Noise measurement; Rails; Box Gaussian; Constrained filtering; Estimation; Rail navigation;

fLanguage

English

Publisher

ieee

Conference_Titel

India Conference (INDICON), 2010 Annual IEEE

Conference_Location

Kolkata

Print_ISBN

978-1-4244-9072-1

Type

conf

DOI

10.1109/INDCON.2010.5712612

Filename

5712612