Key Technology On Asteroids Navigating And Modeling In Deep Space Exploration

Deep space exploration is an important way to help human understanding the universe and discovering space environment. Due to the long flight time, great distance and complex deep space communication environment, deep space exploration based on the control of ground is limited. Therefore, deep space exploration based on optical autonomous navigation is developed, by using a series of sensor carried on bound. Without relying on the control from ground, probes carry out orbit determination and attitude control independently.Most deep space exploration missions are chose asteroids as the target, the detection forms including flyby, impact, landing, sampling and return. In cruise segment, asteroids are usually selected as the observation of navigation, and the screening of navigation ‘beacons’ would affect the accuracy of navigation directly. In approach segment, the target star is the observation of navigation, a nd the extraction of target’s centroid affect tasks carried out in approach segment. To solve the above problems, this paper based on the key technology of optical autonomous navigation in deep space exploration, mainly completed the following three aspect s:(1)Optical navigation for deep space exploration in cruise phase generally select several asteroids as navigation beacon stars. In order to improve the accuracy of navigation, optimal selection of these asteroids are required. In this paper, based on a simulation of Mars exploration mission, combined with the design of Mars orbit, criteria including absolute magnitude of the asteroid with the distance from the sun, the solar phase angle, the distance and line of sight angle between detector and asteroids are given to select the optimal combination of asteroids. In-orbit Secondary Screening Method is proposed. The method demonstrated with mathematical simulation and the analysis of the simulation results indicates that the algorithm can effectively improve the accuracy of navigation.(2)The experiment expense of deep space exploration is much expensive. Scientific research can take by simulation to validate. Based on the simulation of Chang’e-2 flew-by the asteroid of 4179 Toutatis. We analog the extraction of optical observations used for autonomous navigation in approach segment. By adding a certain intensity of Gaussian white noise to the images generated by POV-Ray tracing application, the star map are simulated. After denoising and thresholding, the threshold centroid algorithm and threshold squared weighted centroid algorithm are used to extract the centroid of the target. Comparing the results with the true centroid, the threshold squared weighted centroid algorithm has a higher accuracy.(3)Deep space exploration after the target celestial body at close range imaging, high-resolution images through further processing to get some of the characteristics of the target celestial body. Further process of the high-resolution images taken by the probe when approach the target, the body surface characteristics of the target was recorded. The third work of this paper is based on the optical images taken by chang’e-2 probe during flew-by asteroid 4179. We customized a method of frequency domain data fusion, which combines the topography information of radar model and the 3rd dimension information estimated from optical image by shape from shading algorithm, and gave out a new Toutatis ’ radar model. And then three evaluation criteria are chose to analyze the new m odel and a model with abundant surface characteristics had been resulted.