Research On Application For Soft Landing On Mars Using Image Matching Technology

In Mars soft landing,the spacecraft position is an important parameter to achieve accurate landing,so the location of the spacecraft for Mars soft landing mission research is of great significance.Image matching technology provides a good prospect for the positioning of spacecraft.This paper focuses on the application of image matching technology in the location parameter estimation of Mars spacecraft.Firstly,the features of satellite image and landing camera image are analyzed,and the acquisition method of the landing point satellite image and the simulated generation method of landing camera image for the land landing process are determined.This paper introduces the related research contents of image matching technology,and combines the satellite image and the characteristics of landing camera image to determine the image matching algorithm for the landing of Mars spacecraft.Secondly,an image matching algorithm based on multi-feature fusion is proposed.Aiming at the complex background of Mars image,the feature detector and feature descriptor which is suitable for the Mars image is selected by evaluating the performance of the common feature extraction algorithms.The matching location algorithm proposed in this paper has stronger performance.Based on this,a robust feature descriptor is proposed based on the weighted fusion rule.The simulation results show that the multi-feature fusion matching algorithm can improve the matching accuracy compared with the single feature,and achieves better robustness under complex Mars environment.Thirdly,the algorithm of relative position estimation of spacecraft based on multi-feature fusion is studied.In this paper,the influence of multi-feature and single feature on the performance of fundamental-matrix estimation are analyzed for inter-frame image matching in relative position estimation.Experimental results show that multi-feature fusion matching can improve the accuracy of the inter-frame matching transformation model.Finally,an absolute position estimation algorithm based on phase correlation matching and multifeature fusion is proposed.According to the relationship between the height of the camera and the number of matching points,a hierarchical matching strategy is designed.The simulation results show that the proposed method can solve the problem of matching performance degradation caused by the difference of resolution between the descend image and satellite image as for altitude change.