Research On The Techniques Of Location And Guidance For The Procedure Of Key Lunar Sampling And Encapsulation Based On Photogrammetry

Highly accurate measurement of the position and attitude parameters of the sampler is an indispensable step in guiding the sampler in completing the task of sampling and packaging in the lunar region.In this paper,we propose a scheme of positioning and guiding the sampler by means of visual measurement by using the surveillance camera inherent in the detector,arranging a small number of artificial markers on the detector.However,the image distortion is large due to the lower resolution of the surveillance camera and the larger angle of view.In addition,different from the ground environment,the lighting environment of the lunar environment is complex and seriously affects the location of the sign.In order to overcome the above difficulties,this paper studies the pose estimation method based on the landmark points and the relevant camera measurement techniques to improve the accuracy.The main research contents include the following four aspects:Aiming at the problem of serious camera distortion,this paper presents a camera distortion correction method based on quasi-Newton method.This method solves the nonlinear equation in Brown's aberration model by using quasi-Newton method,and removes the influence of aberration with high accuracy.Compared with the traditional method,this method has a wider convergence area.For the problem of high-precision diagonal mark extraction in complex light environment,the paper preprocesses the image by using the image processing technology such as filtering and binarization.The method of generating the ideal diagonal mark template is proposed,and the template matching The method overcomes the problem of lower camera resolution and larger noise.Through the above-mentioned series of targeted methods,the accuracy of diagonal mark extraction in complex light environment is effectively improved.Aiming at the problem of the influence of the mark arrangement on the measurement accuracy,this paper presents an ideal precision analysis method based on the error propagation matrix,and verifies the accuracy of the method by comparing the calculation accuracy of the method of solving pose and pose.Based on this ideal accuracy,the influence of different marker points layout on the measurement accuracy was studied.Compared with the traditional method of simulation analysis,this method does not need a lot of simulation analysis,has good stability,at the same time more objective.In order to solve the pose estimation problem in complex light environment,this paper presents a robust pose estimation method based on MM estimation.The method introduces the MM estimation method in regression analysis into pose estimation,and automatically determines the outliers by weighted method,and adjusts the weights of outliers to zero.Compared with the traditional PnP method,this method can effectively overcome the effect of outliers and improve the robustness of pose measurement.Based on the above algorithm,this paper also designed and developed a lunar sampling and visual guidance software system.Through the sampling and image processing of the Chang'e-5,the sampling and packaging process was successfully guided and the Chang'e-5 sample and the positive sample-test.The software can effectively assist in the completion of sampling tasks,improve operational efficiency.