基于值域的多约束GNSS单频单历元定姿新算法

多天线增强了全球卫星导航系统（GNSS）单频单历元姿态测量解算模型的强度，但随着天线数量增加，模糊度维数成倍增加，从模糊度域确定搜索空间的计算量将显著增长。基于此，提出了基于值域的多约束多天线GNSS单频单历元姿态测量新算法：该算法将姿态约束融入值域搜索模型，利用姿态约束条件推导搜索步长，通过姿态域三维搜索确定模糊度搜索空间，以基于最优条件姿态解非迭代近似估计的方法固定模糊度。实验结果表明，新算法中模糊度搜索效率较原方法提高约65.8%，固定模糊度效率较标准迭代算法提高约95.3%，且与标准迭代算法性能相当；所提算法能够实现GNSS单频单历元的模糊度固定和载体全姿态测量，具有较高的正确率。

A new multivariate constrained algorithm for GNSS single-epoch single-frequency attitude determination based on attitude domain

The model of Global Navigation Satellite System (GNSS) single-epoch single-frequency attitude determination can be strengthened with multiple antennae. However, the more antennae, the higher the dimension of ambiguity. If we still search ambiguity in the ambiguity domain, the efficiency of search will decrease significantly. To solve this problem, this paper proposes a multivariate constrained algorithm for GNSS single-epoch single-frequency attitude determination based on attitude domain. First, the prior information of pitch and roll is integrated into the search process to derive the search step-length of attitude angle, and the integer candidates is determined by traversal search in the three-dimension attitude domain. Then, the cost function is parameterized with the Euler angle, and non-iterative approximate estimation of optimal conditional attitude solution is utilized to simplify the iterative computation in calculating the cost function. Experimental results show good performance of our new method. By integrating prior information of pitch, the efficiency of search in the attitude domain is improved by 65.8%. Compared to the standard iteration algorithm, non-iterative approximate estimation of conditional attitude solution reduces the time consumption of calculating the cost function by 95.3%, and the method proposed has nearly equal performance to that of the standard iteration algorithm. By the new algorithm, GNSS full attitude determination can be realized with high success rate and reliability.