Scholarlink

<p>In complex indoor environments, traditional localization methods often suffer from non-line-of-sight (NLOS) and multipath problems, which lead to unsolvable or incorrectly solved mathematical localization models, thereby limiting localization accuracy. A localization method based on swarm intelligence optimization has been proposed to address this issue. The swarm intelligence optimization algorithm does not require solving matrix inversions and transforms the localization problem into a function optimization problem, which can obtain approximate optimal solutions. Nevertheless, optimization algorithms are beset with issues like sluggish convergence speed and proneness to getting trapped in local optima, thereby failing to satisfy the current practical requirements for localization. This paper proposes a new method that applies the grey wolf optimization (GWO) algorithm to ultra-wideband (UWB) indoor localization to enhance localization accuracy. It improves the GWO algorithm with four strategies. Firstly, a small-area optimization strategy near the target point is proposed. The Chan algorithm is adopted for initial tag localization, and the initial localization result is taken as a constraint to construct the search area of the GWO algorithm, thereby reducing the large space region to a small space region and enhancing optimization efficiency. Secondly, an improved Tent mapping, a nonlinear convergence factor, a fitness-weighted location update strategy, and an out-of-bounds reflection mechanism are designed to improve the GWO algorithm, referred to as the TIGWO algorithm. Finally, apply the TIGWO algorithm to determine the optimal location of the tag. The experimental results indicate that the proposed algorithm significantly enhances indoor localization accuracy. Compared to the Chan, Chan-Taylor, PSO, WOA, and GWO algorithms, the average localization accuracy has been enhanced by 59.65%, 63.41%, 40.97%, 45.97%, and 35.44%, respectively. In an equipment warehouse scenario, the X-axis, Y-axis, and Z-axis localization errors are 0.129 m, 0.101 m, and 0.154 m, respectively.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

An improved grey wolf algorithm and its localization research in complex indoor environments

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年3月1日

查看全文
  • Bing Li,
  • Yanxin Hao,
  • Yiyang Cui,
  • Xingshao Chai,
  • Jingmei Zhou,
  • Chungang Liu

摘要

In complex indoor environments, traditional localization methods often suffer from non-line-of-sight (NLOS) and multipath problems, which lead to unsolvable or incorrectly solved mathematical localization models, thereby limiting localization accuracy. A localization method based on swarm intelligence optimization has been proposed to address this issue. The swarm intelligence optimization algorithm does not require solving matrix inversions and transforms the localization problem into a function optimization problem, which can obtain approximate optimal solutions. Nevertheless, optimization algorithms are beset with issues like sluggish convergence speed and proneness to getting trapped in local optima, thereby failing to satisfy the current practical requirements for localization. This paper proposes a new method that applies the grey wolf optimization (GWO) algorithm to ultra-wideband (UWB) indoor localization to enhance localization accuracy. It improves the GWO algorithm with four strategies. Firstly, a small-area optimization strategy near the target point is proposed. The Chan algorithm is adopted for initial tag localization, and the initial localization result is taken as a constraint to construct the search area of the GWO algorithm, thereby reducing the large space region to a small space region and enhancing optimization efficiency. Secondly, an improved Tent mapping, a nonlinear convergence factor, a fitness-weighted location update strategy, and an out-of-bounds reflection mechanism are designed to improve the GWO algorithm, referred to as the TIGWO algorithm. Finally, apply the TIGWO algorithm to determine the optimal location of the tag. The experimental results indicate that the proposed algorithm significantly enhances indoor localization accuracy. Compared to the Chan, Chan-Taylor, PSO, WOA, and GWO algorithms, the average localization accuracy has been enhanced by 59.65%, 63.41%, 40.97%, 45.97%, and 35.44%, respectively. In an equipment warehouse scenario, the X-axis, Y-axis, and Z-axis localization errors are 0.129 m, 0.101 m, and 0.154 m, respectively.