A novel 3D feature detection and matching approach for autonomous planetary landing mission. (May 2019)

• Record Type: Journal Article
• Authors: Tian, Yang; Yu, Meng; Zhang, Yingying
• Is Part Of: Proceedings of the Institution of Mechanical Engineers. Volume 233:Number 6(2019)
• Page Start: 2241
• Page End: 2249
• Publication Date: 2019-05
• Subjects: Distance set -- feature description -- feature detection -- harmonic coefficient -- planetary landing; Aeronautics -- Periodicals; Astronautics -- Periodicals; Airplanes -- Design and construction -- Periodicals; Aerospace industries -- Periodicals; 629.1
• DOI: 10.1177/0954410018774696
• Languages: English
• ISSNs: 0954-4100

Abstract

In an autonomous planetary/asteroid landing mission, landmark recognition is crucial to the success of the navigation system. The failure of feature detection or matching could lead to evident increase of bias in lander pose estimation. To this end, we propose a novel 3D feature detection and matching algorithm in this paper. The spherical harmonic coefficients are adopted to describe a 3D natural feature, and a relative distance set feature description approach is proposed as a supplement feature descriptor to enhance the distinctiveness of 3D feature. Simulation results demonstrate the effectiveness of our complete feature detection and matching algorithm in terms of feature detection rate and correct feature matching rate.