一种基于改进天牛须探索算法的多点定位算法

A Multi-point Positioning Algorithm Based on Improved Beetle Antennae Search Algorithm

天牛须探索(Beetle Antennae Search,BAS)算法具有搜索速度快、运算量少和实施便捷等优点,受到越来越多研究者的关注。但是,由于天牛的大小,BAS算法并不适合远距离的定位,这限制了BAS算法的进一步应用,同时,天牛每次移动的步长未能随着算法的运行而实时变化,这将会导致天牛每次移动的距离与定位所需存在一定的不适应性。针对这种情况,对BAS算法进行改进,提出了基于Chan算法的改进BAS算法。通过对BAS算法各个步骤的分析,针对天牛初始位置、天牛迭代运行方程以及天牛每次移动步长结合实际情况进行改进。对初始位置采用一次定位方式进行确定,将大空间区域定位缩小为小空间区域;对运行方程增加一个实时运行角度进行实时变化,将每次移动距离由定值转换为变值;对步长采用训练方式进行确定,将最合适的步长应用于定位。再将所得数据运用到Chan算法中进行定位。经过Matlab仿真可以发现,经过改进后的BAS算法相较于之前有很大的优化。

The Beetle Antennae Search(BAS)algorithm has advantages such as fast search speed,less calculation,and convenient implementation,which has attracted more and more researchers’attention.Due to the size of the beetle,however,the BAS algorithm is not suitable for long-distance positioning,which limits its further application.At the same time,the step of each movement of the beetle fails to change in real time with the operation of the algorithm.This will lead to a certain incompatibility between the distance the beetle moves and the distance required by positioning.To address this issue,the BAS algorithm is improved,and an improved BAS algorithm based on Chan algorithm is proposed.Through an analysis of each step of the BAS algorithm,the initial position of the beetle,the iterative running equation of the beetle and the step length of each movement of the beetle are improved in combination with the actual situation.The initial position is determined by one-time positioning,which can reduce a large space area of positioning to a small one.A real-time running angle is added to the running equation to realize real-time variation,which can convert each moving distance from a fixed value to a variable value.And the step length is determined by training,and the most appropriate step length is used for positioning.Finally,the obtained data is applied to Chan algorithm for positioning.The results of Matlab simulation experiments show that the positioning effect of the improved BAS algorithm is greatly optimized compared to the previous ones.