基于改进PSO的两点磁梯度张量定位方法

Two point magnetic gradient tensor localization method based on improved PSO

摘要:

针对磁梯度张量定位法中单点定位受地磁场影响误差大、多点定位优化算法中对局部最优解敏感及定位精度差等问题,提出了基于改进粒子群优化算法的两点磁梯度张量定位方法。该方法基于磁偶极子理论求解磁梯度张量分量；并利用两点信息建立目标位置和磁梯度张量之间的非线性目标函数,最后采用基于动态调整的粒子群算法对目标位置坐标进行求解,通过改进惯性权重和学习因子,使其从固定值变为随搜索过程中适应度函数变化而非线性变化的值。结果表明,与单点定位法及传统粒子群算法的两点定位法相比,该方法将平均定位误差分别由35.16 cm、12.6 cm降低至5.58 cm,明显减小了定位误差,且该方法具有受地磁场影响较小、自动平衡全局搜索和局部搜索及抗噪性等优点。

Abstract:

In response to the problems of large errors in single point positioning due to the influence of the geomagnetic field in the magnetic gradient tensor positioning method, sensitivity to local optimal solutions in multi-point positioning optimization algorithms, and poor positioning accuracy,a two-point magnetic gradient tensor localization method based on improved particle swarm optimization algorithm was proposed. This method is based on the magnetic dipole theory to solve the magnetic gradient tensor components.And use two-point information to establish a nonlinear objective function between the target position and the magnetic gradient tensor.Finally, the target position coordinates are solved using a particle swarm algorithm based on dynamic adjustment.By improving the inertia weight and learning factor, it is changed from a fixed value to a value that varies nonlinearly with the fitness function during the search process.The results show that compared with the single point positioning method and the traditional particle swarm algorithm"s two-point positioning method, this method reduces the average positioning error from 35.16 cm and 12.6 cm to 5.58 cm, respectively, significantly reducing the positioning error. Moreover, this method has the advantages of being less affected by the geomagnetic field, automatically balancing global and local search, and noise resistance.