Measurement Domain-based Real-time Multiantenna Gnss Attitude Determination Considering Baseline Deformation

In recent years,with the rapid development of Global Navigation Satellite System(GNSS),real-time attitude determination based on multi-antenna GNSS has become a research hotspot in attitude determination field due to its high precision,all-weather,long-endurance and short initialization time.In order to improve the performance of attitude determination effectively,this thesis compares Position domain Attitude Determinationalgorithm(PAD)with Measurement domain Attitude Determination algorithm.Due to the influence of baseline deformation,the attitude determination performance of Measurement domain Attitude Determination algorithm(MAD)is decreased.A new algorithm-Enhanced Measurement domain Attitude Determination algorithm is proposed to solve the problem.In view of the high dimension of ambiguity caused by multiple baselines,different methods for solving ambiguity are compared and analyzed,and a low-cost attitude determination system is built.The main research contents of this article are as follows:(1)In order to achieve the high-precision performance of attitude determination,traditional GNSS attitude determination algorithms are studied.Position domain Attitude Determination algorithm calculates attitude information by solving the baseline vectors in the Earth-Centered Earth-Fixed coordinate system.Measurement domain Attitude Determination algorithm uses the transformation relationship of the baseline vectors between different coordinate systems to compute all attitude angles while solve the ambiguity.The mathematical models of traditional GNSS attitude determination algorithms are derived.The Experiments show that compared with Position domain Attitude Determination algorithm,Measurement domain Attitude Determination algorithm improves the roll angle accuracy and the continuity of attitude determination.When the number of satellite is limited,continuity improvement is more obvious.Therefore the Measurement domain Attitude Determination algorithm has stronger applicability in practical application.(2)According to the problems existing in the Measurement domain Attitude Determination algorithm,an improved algorithm is proposed.Since Measurement domain Attitude Determination algorithm needs accurate prior baselines information,and the carrier is not a completely rigid-body,the movement of the carrier may result in baseline deformation,so that the performance of algorithm maybe decline.Aiming at this problem,this thesis proposes a new Measurement domain Attitude Determination algorithm that takes into account baseline deformation-Enhanced Measurement domain Attitude Determination algorithm.It calculates the corresponding covariance matrix by estimating the baseline deformation to obtain an accurate random model,which is more accurate than the mathematical model of Measurement domain Attitude Determination algorithm.The results of simulation show that Enhanced Measurement domain Attitude Determination algorithm is up to 96.9% higher than that in Measurement domain Attitude Determination algorithm.(3)The Measurement domain Attitude Determination algorithm employs a combination of multi-baseline,multi-system,and multi-frequency observations,which leads to too high ambiguity vector dimension.There may also be problems that the new algorithm proposed in this article causes the accuracy of the floating-point solution to decrease,and errors may occur.This dissertation researches full ambiguity solution and partial ambiguity solution,which deduces Least square AMBiguity Decorrelation Adjustment resolution and Partial Ambiguity Resolution based on that the MSRR(Minimum Success Rate Requirement)criterion.The experiment shows that when three baselines are used in the static experiment,the continuity of Measurement domain Attitude Determination algorithm using Partial Ambiguity Resolution,is improved by 3.2% compared with Measurement domain Attitude Determination algorithm using Least square AMBiguity Decorrelation Adjustment resolution.When dynamic experiments use two baselines,the continuity of Measurement domain Attitude Determination algorithm using Partial Ambiguity Resolution is improved by 2.6% compared with Measurement domain Attitude Determination algorithm using Least square AMBiguity Decorrelation Adjustment resolution.Enhanced Measurement domain Attitude Determination algorithm using Partial Ambiguity Resolution is improved by 0.5% compared with Measurement domain Attitude Determination algorithm using Least square AMBiguity Decorrelation Adjustment resolution.Therefore,the Partial Ambiguity Resolution is more effective when the dimension of ambiguity is higher,and the algorithm proposed in this paper can obtain the best performance based on the Partial Ambiguity Resolution.(4)This thesis completes a set of low-cost GNSS attitude determination system by building a hardware platform and writing software,and tests the system experimentally.The test found that in a general environment,the low-cost GNSS attitude determination system has a continuity of ?90%,a roll angle accuracy of ?0.5°,and a pitch angle and a heading angle accuracy of?0.2°..