摘要:
在BDS/GPS组合定位中,选择空间位置最优的卫星组合是第一步,也是十分重要的一步.传统选星算法在选取最佳卫星组合过程中涉及大量的矩阵乘法和求逆运算,计算量大,定位实时性低.为解决快速选星定位问题,综合考虑定位精度和实时性要求,提出一种新的选星算法,该算法将BP神经网络和遗传算法相结合,并将几何精度因子(GDOP)作为判断定位精度高低的依据.将应用该算法得到的GDOP和运算时间与最小几何精度因子法所得对应结果进行比较,实验结果表明,该算法在大大减小了计算量的同时保证了定位精度,具有良好的实时性和可行性.
Abstract:
In BDS/GPS combination positioning,it is a very important step to select the satellite combination with the best spatial location.The traditional satellite selection algorithm involves a large number of matrix multiplication and inversion operations,so the calculation is large and the real-time is low.For the problem of rapidly fixing position,the authors,considering the positioning accuracy and real-time requirements,propose a new satellite selection algorithm,which combines the BP neural network and genetic algorithm,and uses the geometric dilution of precision (GDOP) as the basis of judging positioning accuracy.Through the comparison of GDOP and the running time acquired by this algorithm and the method of minimum geometric dilution of precision,it is found that the proposed algorithm can greatly reduce the computational complexity and ensure the positioning accuracy,thus exhibiting good real-time and feasibility.
下载: