一种分步组合的部分模糊度固定方法

A sub-step combination method of partial ambiguity resolution

针对高精度定位中模糊度固定率与正确率低的问题,提出了一种分步组合的部分模糊度固定方法。该方法首先采用丹麦法对双差观测值随机模型进行调整,并依据标准化残差选出相位观测值明显异常的模糊度子集;然后基于降相关模糊度方差法进行精筛选;同时顾及模糊度浮点解信息,优先剔除最优和次优解不一致的模糊度;采用最小二乘降相关平差(LAMBDA)法进行固定。实验结果表明:相对于全模糊度固定法,部分模糊度固定法可有效地提升模糊度固定率,提升了大概13.8%~37.0%;分步组合的部分模糊度固定法,有效减小了实际观测值中未探测到偏差对模糊度固定的影响,对于部分观测值异常的场景也能较好适用,较原始的部分模糊度固定法,动态场景下平均改善61.5%,并提升了模糊度固定解的正确率。

Aiming at the problem of low accuracy and fixed rate of ambiguity in high precision positioning,a sub-step Combination Partial Ambiguity Resolution(C-PAR)method is proposed.Firstly,the Danish method is used to adjust the random model of double difference observations,and the ambiguity subset with obvious abnormal phase observations is selected according to the standardized residual;Then,the method of reduced correlation ambiguity variance was used for fine screening;At the same time,considering the ambiguity information of floating point solution,the ambiguity inconsistency between optimal and suboptimal solution is eliminated first;The Least-square Ambiguity Decorrelation Adjustment(LAMBDA)method was used to fix them.The experimental results show that,compared with the Full Ambiguity Resolution(FAR)strategy,the Partial Ambiguity Resolution(PAR)strategy can effectively improve the ambiguity fixed rate,and the improvement degree is about 13.8%-37.0%;The C-PAR method can effectively reduce the influence of undetected deviation in the actual observation value on the ambiguity resolution,and can also suitable for some scenes with some abnormal observation value,compared with the original partial ambiguity resolution method,the average improvement in dynamic scenes is 61.5%,and the accuracy rate of ambiguity fixed solution is improved.