Emergency Return Method of Lunar Rover Based on Rut Tracking

Lunar rover performs inspection and exploration missions in a complex unstructured environment. When the astronaut is in an emergency situation, the astronaut does not have the ability to control lunar rover, and lunar rover needs to independently track and navigate according to environment informat...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:2025 4th Conference on Fully Actuated System Theory and Applications (FASTA) s. 2776 - 2780
Hlavní autori: Zheng, Bo, Hu, Tao, Huang, Fei, Qian, Zhouyuan, Zhang, Hanmo, Cao, Tao
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: IEEE 04.07.2025
Predmet:
Accuracy
Autonomous navigation
Emergency return
Feature extraction
Fitting
LSD algorithm
Lunar rover
Moon
Navigation
Rut tracking
Soil
Space vehicles
Surface texture
Transforms
Wheels
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Lunar rover performs inspection and exploration missions in a complex unstructured environment. When the astronaut is in an emergency situation, the astronaut does not have the ability to control lunar rover, and lunar rover needs to independently track and navigate according to environment information and rut information to achieve reliable emergency return. Lunar surface is characterized by strong light, strong shadow and weak texture. Meanwhile the limited contact force between wheel and soil caused by microgravity and inconsistency of soil softness on lunar surface all lead to the difficulty of rut detection and extraction. Rut information is an obvious artificial landmark feature of lunar surface, which can enrich the navigation source and improve the completeness and accuracy of navigation. Therefore, an emergency return method of lunar rover based on rut tracking is proposed in this paper. Firstly, the improved LSD (Line Segment Detector) algorithm is used to screen rut concentration area. Secondly, sobel convolution kernel is used to detect image edge. And then transverse rut distribution is obtained by quadratic Gaussian filtering and binary processing. Vertical rut distribution is obtained by using vertical feature fitting and Hough transform. Finally, the middle line of rut is extracted by fitting. The simulation results show that the proposed method can effectively extract rut and accurately identify middle line of rut. Meanwhile the navigation accuracy is improved by more than 1%.
DOI:10.1109/FASTA65681.2025.11138619