Research On Optical Navigation Of Mars Descent And Landing Phase

In human deep space missions, Mars exploration is a hot point because of its great scientific significance. In order to obtain more valuable scientific data, presicion landing on regions with abundant ladscape and sampling and returning to earth is the development trend of Mars missions. But according to the present landing missions, their landings accuracy didn’t reach the requirements of precision landing. Navigation system is the eye of the landers control system. Its presicion plays a key role for lander’s eventually lading precision. Therefor, it is necessary to improve the navigation system for landing precision.This thesis considers powered landing stage and final landing stage of the Mars exploration task as the background. The navigation scheme is designed according to the characteristics of the two stages. The related influencing factors of the navigation accuracy is analyzed and verified by the way of mathematical simulation. At the same time, this thesis has had a research of selecting characteristic points aimed at the powered landing stage. The main research contents of the thesis as follows:Firstly, autonomous navigation method of landing stage based on features of the known locations has been researched. Provided that the characteristic location has been acquired, the thesis researches methods of fusing inertia and visual navigation information to estimate the landing state. The definition of commonly used coordinate system of the powered landing stage is given in this thesis and dynamic and measurement models of lander in powered landing stage in Reference Frame are built. Taking nonlinearity of the system into consideration, information about landing state has been estimated according to the metrical information of filter inertial fusion measurement unit designed according to the expanded Kalman Filtering algorithm and the characteristic location information. Various factors’ influence on navigation accuracy has been analyzed by method of mathematical simulation.Secondly, method of selecting characteristic points in powered landing stage based on entropy has been researched. Concept of entropy is introduced and the definition of entropy as well as its probabilistic relationship with random variable is introduced according to development of information theory. Methods of selecting characteristic points in powered landing stage based on entropy is set with combination of entropy theory and the navigation filtering estimator in powered landing stage preciously established. Algorithmic accuracy is analyzed and verified by mathematical simulation.At last, autonomous navigation of the multi-information fusion in final landing stage is researched. Dynamic model is built; The autonomous navigation scheme that the information is observed with the combined effect of laser range finders, navigation camera and inertial element has been put forward according to characteristics of navigation tasks in this thesis. Navigation filter has been designed according to EKF algorithm. Various factors’ influence on navigation accuracy has been analyzed by method of mathematical simulation.