基于方差分量估计的多模GNSS/声学联合定位方法

Multi-GNSS/Acoustic joint positioning method based on variance component estimation

由于海洋环境的特殊性,现阶段通常采用卫星定位与声脉冲测距相结合的方式建立海洋大地测量控制网。针对BDS/声学联合定位海底点时存在收敛慢、精度受限等问题,提出融合BDS、GPS、Galileo多系统观测数据的多模GNSS/声学联合定位方法,并进一步引入Helmert方差分量估计,以改善因随机模型不准确而影响融合定位性能的情况。采用海上实测数据和模拟水下测距数据进行了实验验证,结果表明:较之单BDS/声学定位模式,多模GNSS观测数据的加入可以加速组合滤波收敛,提高定位精度和稳定性;引入Helmert方差分量估计精化随机模型后,双系统模式下三维点位误差RMS值约为0.2m,三系统模式下三维点位误差RMS值约为0.1m,联合定位性能得到进一步提升。

Due to the specificity of marine environment,the combined way of satellite positioning and acoustic ranging is usually adopted to establish the marine geodesic control network.To solve the existence problem in BDS/acoustic joint positioning such as slow convergence and limited accuracy,a multi-GNSS/ acoustic joint positioning method based on the fusion of BDS,GPS and Galileo multi-system observation data is proposed,and then the Helmert variance component estimation is further introduced,thus the phenomenon that fusion positioning performance would be influenced by inaccurate stochastic model can be improved. The measured data on the sea and the simulated underwater ranging data are used to carry out the experiment. Results show that,compared with BDS/acoustic positioning method,the addition of multi-GNSS observation data can accelerate the convergence of combined filtering and improve the accuracy and stability of positioning.After introducing the Helmert variance component estimation to refine the stochastic model,the RMS value of 3D point bias in the dual-system mode is about 0.2 m,and the value is about 0.1 m in the three-system mode,showing that the performance of joint positioning is further improved.