Advanced Search
WANG Jia, LI Dafei, HE Ximing, CHENG Ziqing, XU Qian, QIAN Xueru, WAN Wenhui. High Precision Localization of Zhurong Rover Based on Multi-source Images[J]. Journal of Deep Space Exploration, 2022, 9(1): 62-71. DOI: 10.15982/j.issn.2096-9287.2022.20210124
Citation: WANG Jia, LI Dafei, HE Ximing, CHENG Ziqing, XU Qian, QIAN Xueru, WAN Wenhui. High Precision Localization of Zhurong Rover Based on Multi-source Images[J]. Journal of Deep Space Exploration, 2022, 9(1): 62-71. DOI: 10.15982/j.issn.2096-9287.2022.20210124

High Precision Localization of Zhurong Rover Based on Multi-source Images

  • Zhurong rover landed on the southern part of Utopia Planitia of Mars on 15 May, 2021(UTC+8), marking that China successfully fulfilled the goals of “orbiting, landing on and roving around” Mars on its own for the first time. Localization of the rover is critical for supporting science and engineering operations in planetary rover missions, such as rover traverse planning and hazard avoidance. This paper introduced the localization method of Zhurong rover based on multi-source images in detail. Based on the coarse-to-fine strategy, the accurate landing positioning was achieved in the Digital Orthophoto Map (DOM) generated by the high resolution orbiter-image. Using the ground correction method combining relative localization and absolute localization, high-precision continuous localization of the rover was realized. Simulation experiment shows that the localization accuracy of the lander was within a pixel of the DOM generated by the high resolution orbiter-image, and the relative localization accuracy based on the Bundle Adjustment (BA) was better than 3% when the distance was about 10 meters. The methods have been successfully applied to the localization of Zhurong rover. The high-precision localization results greatly support the rover’s efficiently roving on Mars surface and avoiding potentially dangerous regions.
  • loading

Catalog

    Turn off MathJax
    Article Contents

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return