Font Size: a A A

Research On GNSS-based Attitude Determination Of High Earth Orbit Satellites

Posted on:2014-04-03Degree:MasterType:Thesis
Country:ChinaCandidate:S L WangFull Text:PDF
GTID:2252330422452868Subject:Navigation, guidance and control
Abstract/Summary:PDF Full Text Request
The high earth orbit satellites are playing an increasingly important role in the field of nationaldefense and civilian. Currently the attitude determination of HEO satellite is achieved mainly throughthe attitude sensors like, gyroscopes, star sensors. However, these sensitive devices exist easy to wearand be disturbed defects. Therefore, the research on the use of satellite-borne GNSS receivers forattitude determination has become a new research direction. In this paper, the main research is on theinitial single difference integer ambiguity solution method, the cycle slip detection and inspectionmehods, and the algorithms for attitude determination of the HEO satellite.Firstly, the dissertation introduces an overview about GNSS and the development of thenavigation technology based on GNSS. By using the STK simulation, the visibility of satellite-borneGNSS receivers is analyzed. Then the article gives the observation models of GNSS attitudedetermination systems, summarizes the principles of multi-frequency phanse combinations, analyzesthe error propagation of the combination process, provides the theory basis for the subsequent work.The methods that are used to solve initial single difference integer ambiguity for satellite-borneGNSS receiver are studied. As the TCAR algorithm has the selection problem of wide lanecombination coefficient, the article gets the optimal coefficient by establishing a probability ofsuccess performance indicators. When the number of GNSS system’s visible stars is below3epoch,TCAR algorithm can not verify the correctness of solving numerical. An improved TCAR methodcombined with recursive weighted least squares algrorthm is proposed to this problem. When thenumber visible stars is not less than3, in accordance with probabilistic characteristics, by giving theimproved TCAR algrorithm a search space and using the baseline constraint, the search algorithm canfaster determine of the single diference ambiguity. Eventually the satellite-borne GNSS receiver cansolve the single diference ambiguity under any visible stars.Using the multi-frequency data to detect and repair the cycle slip is studied. By expansing andimproving the traditional dual-band MW algorithm to tri-band, an improved algthm is proposed. Then,the paper analyzes the cycle slip detection test volume and test threshold selection problem in thethree systems. By construting three linearly independent combination coefficient inspections, theimproved method can detect any cycle slips on GNSS three combination carriers, and then restore thehop count to basic carriers, repair the cycle slips.The deterministic algorithms and estimation algorithm for GNSS attitude determination systems are studied. When the visible of satellites are not less3, the article analyzes and compares the satelliteattitude accuracy of QUEST algorithm, in the case of single system, dual-system and multi-system,with single-frequency and multi-frequency. But the deterministic algorirhms and nomal filteringalgorithm can not solve the attitude, when the visible stars are below3. The paper also studies thepredictive filitering algorithm’s attitude solving situation in single system, dual-system andmulti-system case, and then compares the accuracy with QUEST algorithm. An improved algorithmby fusing the two algorithms is proposed to improve the speed and accuracy of the attitude solution.Finally, by establishing a multi-constellation system in HEO satellite attitude simulation platform, thisdissertation simulates a period of attitude trajectory and verifies the correctness of the aboveresearchs.
Keywords/Search Tags:Mulit-consatellation Integrated System, High Earth Orbit Satellites, Visibility, Multi-frequency Carrier, Interger Ambiguity, Success Rate, Cycle Slip, Attitude Dertermination, QUEST, predictive filtering
PDF Full Text Request
Related items