附加速度与基线长约束的海上基线网平差

A maritime baseline vector network adjustment method with additional velocity and baseline length information constraints

为削弱全球卫星导航系统(GNSS)测量误差对海底控制点测量的影响,对GNSS基线解算后的数据进行后处理,并根据海上作业环境提出一种基于速度和基线长等先验信息的约束卡尔曼滤波(KF)方法:充分利用海上多天线的观测模式,将接收机间的速度与距离关系作为约束引入到模型当中,以提高状态估计的精度;然后对GNSS差分定位中基线解算之后的数据进行网平差;最后与最小二乘法和传统卡尔曼滤波进行比较。实验结果证明,与前2种方法相比,提出方法的观测噪声平滑能力更为突出,所选3条基线的基线长偏差的标准差和均方根较RTKLIB软件动态基线解算后的数据分别下降了27%、48%和67%以上;由于水声换能器的三维坐标由GNSS天线坐标转化而来,所以网平差后的GNSS坐标能够提供更加精确的水声换能器坐标,进而提高海底控制点的量测精度。

In order to weaken the influence of global navigation satellite system(GNSS)measurement error on seafloor control point measurement,the paper proposed a constrained Kalman filtering(KF)method based on prior information such as velocity and baseline length after post-processing baseline data under offshore operating environments:the observation mode of multiple antennas at sea was fully utilized,and the relationship between the speed and distance between receivers was introduced into the model as a constraint to improve the accuracy of state estimation;then the net adjustment for the data after baseline solution in GNSS differential positioning was carried out;finally,it was compared with least square(LS)method and traditional Kalman filtering.Experimental result showed that compared with the previous two methods,the proposed method would have a more prominent smoothing ability,and the standard deviation and root mean square of baseline long deviation of the three selected baselines could be decreased by more than 27%,48%and 67%,respectively,compared with the data after dynamic baseline solution by RTKLIB software;moreover,because the three-dimensional coordinates of the underwater acoustic transducer are converted from GNSS antenna coordinates,the GNSS coordinates after net adjustment could provide more accurate coordinates of the underwater acoustic transducer,thus improving the measurement accuracy of the seafloor control points.