In this chapter, three different localization methods have been presented. The first, the voting-based algorithm, provides absolute position measurements and combines simplicity and reliability. Angular distortions affecting the acoustic data are reduced by integrating compass measurements. The second method relies on an Extended Kalman Filter (EKF) for position estimation. Its strong point is the use of the filter itself, which makes it possible to constantly estimate the vehicle's state and merge information from different sensors. This third approach to the localization problem is an attempt to merge the best from the two methods presented. The proposed hybrid method includes an EKF update with the measurements from a Doppler Velocity Log (DVL), a compass and an adapted version of the presented voting algorithm. The benefits are twofold. First, the position estimate will now be corrected with absolute measurements, making it more reliable and avoiding eventual track losses. Second, the position estimates from the filter can be included in the voting process to avoid the dispersion of votes along the vehicle's trajectory while moving.


    Access

    Access via TIB

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Localization with an a priori Map


    Contributors:
    Ribas, David (author) / Ridao, Pere (author) / Neira, Jose (author)


    Publication date :

    2010


    Size :

    29 Seiten





    Type of media :

    Article/Chapter (Book)


    Type of material :

    Print


    Language :

    English