Research On Deep Space Detection Autonomous Navigation Method Based On Nonlinear Filtering

Because of the advantages on the improvement of explorer’s self-depended viability and reduce the reliance to ground stations, autonomous navigation has become one of the key technologies in the deep space exploration. Celestial navigation is a common method of autonomous navigation. Traditional navigating method is base on celestial angles measurement, which has the problem of complicate and huge computation. However, in recent years, a new navigation method based on the Celestial body’s red-shift measuring velocity for navigation was proposed, which had the advantage of high autonomous, no delay and easy to realize, but the method is still in theoretical exploration stage. Besides, hardware, optimization and selection of the navigation filters are also key factors for the improvement of performance. The optimization of filtering algorithm can be given by studying the factor effect system performance of filtering algorithm. At the same time, assessing the filtering performance of navigation system is helpful to select an algorithm for specific navigation system.With the background of Mars exploration mission in transfer orbit, the thesis researches the relationship between the nonlinear filtering algorithm and navigation performance, and assesses the performance of different filtering algorithms. For analysis, we set up a simulation platform using MATLAB. A new navigation system base on spectral red-shift is tested, and the navigation performance under different embedded nonlinear filtering is also evaluated by the module. The effectiveness of the navigation performance will be shown as a numerical or graphical form. Contributions of the thesis are as follows:(1) On public orbit data, we use STK to simulate transfer orbit and generate the calibrated orbit data, which is more an effective approach to the real environment than before.(2) Investigate the methods of autonomous navigation in deep space exploration. We set up a new autonomous navigation MATLAB simulation system using the red-shift navigation method.(3) Apply some filtering algorithms, such as EKF, UKF, UPF, to the new autonomous navigation system, we analyze the factors which effect the system performance of the filtering algorithms, and give the principle to optimize navigation performance.(4) Show the navigation performance under a set of indicators, which provides a reference to optimization and selection of filtering algorithms.