Font Size: a A A

Study On Three Fields Of View Celestial Positioning And Orientation With Minimum Loss Function

Posted on:2016-11-10Degree:DoctorType:Dissertation
Country:ChinaCandidate:L WuFull Text:PDF
GTID:1228330461972957Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
Celestial navigation device uses the exact location of celestial bodies as a reference,it provides accurate position, orientation or attitude information for carrier platform. Due to its advantages of high precision, good concealment and strong autonomy, it’s widely used in the space and near-Earth platform. The three fields of view(FOVs) celestial positioning and orientation system is the branch of celestial navigation technology in the near-Earth platform applications.The main research contents are as follows:Since the positioning and orientation algorithm based on space analytic geometry method has the problem that the orientation precision is affected by the positioning error, A celestial positioning and orientation algorithm with the minimum loss function is proposed. The main idea of this method is to obtain the least squares optimal solution. The accurate positioning and orientation can be solved simultaneously. Experimental results indicate that compared with the traditional algorithm, the positioning and orientation result with the new algorithm is improved by about 26% and 60% respectively, and the result is more stable.The various error sources which affect accuracy of positioning and orientation are pointed out. The characteristic and the probability distribution of error source are summarized; and the impact of error sources on positioning and orientation information pair is assessed. Then, the positioning and orientation error analysis model is established and the error simulation is analyzed by using Monte Carlo method. The thesis points out that dominant error source is horizontal measurement error, followed by data error of deviation of vertical.Compare with the three FOVs positioning and orientation system and the single FOV system on the properties of accuracy and reliability. The thesis points out that the large characteristics database between fields is the bottleneck of the three FOVs star pattern recognition. According to the properties of three FOVs celestial navigation system, gives the three FOVs star pattern recognition principle and guide star selection principle between fields.According to the guide star selection principle between fields, the 1st order self-organizing guide star selection algorithm based on the principle of magnitude is proposed. the guide star data base between FOVs is established, and the number of the guide stars which is selected by this method reduces 48.9% compared with the visual magnitude threshold method. The guide stars distributes evenly. The characteristics database between fields reduces 73.1%. Star pattern recognition simulation tests show that the speed of the star pattern recognition using inter-field navigation catalog upgrade about 41%.According to the star pattern recognition principles, the star pattern recognition strategy which is suitable for the three FOVs system is proposed, the strategy can be divided into four parts including levels of recognition, search method, result inspection and the completion standard. The success rate of all sky star pattern recognition using this strategy is 96.63%, and the recognition time is about 22.5ms. In order to quickly navigate with the known rough azimuth and location information by three FOVs positioning and orientation device,a fast recognition method for local star pattern recognition is proposed.The success rate of local star pattern recognition is improved up to 99.79% and the recognition time is about 18.7msThe outfield experimental system of the three FOVs celestial positioning and orientation is established. Through the analysis of the visible band guide stars distribution, in order to improve the efficiency of star pattern recognition, the optimized optical system parameters is determined. The outfield experimental results show that the average positioning error of the system is 187.7m, the average deviation is 44.1m and the mean orientation error is 2.8 ", the average deviation is 1.6."Finally, the development prospects, technical problems to be solved, the engineering problems of the three FOVs celestial positioning and orientation system are discussed.
Keywords/Search Tags:celestial navigation, positioning and orientation, star pattern recognition, Error analysis, three FOVs
PDF Full Text Request
Related items