This paper develops two algorithms that provide relative navigation measurements for automated aerial refueling solely from a stereo image pair. Algorithms were analyzed in simulation and then in flight test using two C-12C aircraft. The first algorithm, the Vision and Bayesian Inference Based Integrity Monitor (V5), uses Bayesian inference and template matching to return a probability mass function (PMF) describing the position of an observed aircraft. This PMF provides a relative position estimate as well as a protection level, thus providing a degree of navigation integrity. Using both simulation and flight test data, mean V5 spherical error was less than one meter and protection levels reliably characterized algorithm uncertainty. The second algorithm, relative pose estimation with computer vision and iterative closest point (R7), uses stereo vision algorithms and the iterative closest point algorithm to return relative position and attitude estimates. Using both simulation and flight test data, mean R7 spherical error was less than 0.5 meters. Additionally, in flight test, mean R7 attitude errors were less than 3 in all axes.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    Title :

    Integrity Monitoring For Automated Aerial Refueling: A Stereo Vision Approach


    Contributors:

    Publication date :

    2018


    Size :

    352 pages


    Type of media :

    Report


    Type of material :

    No indication


    Language :

    English