Optics and Precision Engineering, Volume. 27, Issue 2, 479(2019)

Attitude accuracy analysis of PnP based on error propagation theory

QU Ye-pin* and HOU Wang
Author Affiliations
  • [in Chinese]
  • show less

    Based on the error relation of a group of typical characteristic points, the most favorable function relation between indirect measurement and direct measurement was established, and the error influence of other characteristic points was synthesized according to the error propagation theory. Finally, a new method of error step-by-step analysis of the PNP problem with complete azimuth, pitch, and tilt angle error analysis model was proposed. Taking the P4P problem as an example, the analytical formulas of the error functions of the related parameters and variables in monocular vision measurement were derived, and the error rules affecting the attitude measurement were revealed. The correctness of the error mathematical model and the validity of the step-by-step error analysis method based on error propagation theory were verified by the simulation of P4P attitude calculation. Under the determined measurement parameters of monocular vision condition, the error analysis model shows that the azimuth angle measurement error is independent of the azimuth angle value, and the error is proportional to the ratio of camera height to cooperative mark size. The pitch/tilt angle has little effect on the measurement error in a larger range. The pitch/tilt angle measurement error is related to pitch/tilt angle value, and the error is proportional to the square of the ratio of camera height to cooperative mark size. The measurement error of the azimuth angle is less than that of the pitch/tilt angle. The analysis method and error function analytical formulas can provide guidance for the design of a monocular vision measurement system.

    Tools

    Get Citation

    Copy Citation Text

    QU Ye-pin, HOU Wang. Attitude accuracy analysis of PnP based on error propagation theory[J]. Optics and Precision Engineering, 2019, 27(2): 479

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category:

    Received: Sep. 3, 2018

    Accepted: --

    Published Online: Apr. 2, 2019

    The Author Email: Ye-pin QU (qypin@126.com)

    DOI:10.3788/ope.20192702.0479

    Topics