Scholarlink

<p>The robotization of rapid roadway excavation equipment is a critical trend in advancing intelligent mining. However, environmental perception in excavation roadways still relies heavily on manual close-range measurements, and effective methods for autonomously acquiring workspace information during rapid excavation remain lacking. Addressing the constraints of low illumination, confined space, and limited field of view in coal mine excavation roadways, this study employs a dual-station laser synchronous scanning system integrated with excavation equipment to capture roadway environmental data and proposes an autonomous point cloud data fusion method. First, initial pose alignment is achieved by matching keypoints between the two station point clouds. Next, the optimal overlapping feature region is dynamically extracted using mutual neighborhood search combined with regional clustering weighting, thereby establishing an adaptive extraction model for this region. Finally, the dynamic feature-optimized iterative closest point (D-FOICP) algorithm is employed to converge the error and complete data fusion. Experimental results demonstrate that the proposed method achieves high-precision point cloud fusion without coordinate guidance or artificial targets, reducing the root mean square error (RMSE) by over 40% and achieving a feature retention rate (FRR) of 96%, providing effective technical support for the autonomous acquisition of operational environmental information by rapid roadway excavation equipment.</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

Workspace information acquisition for rapid roadway excavation equipment using laser scanning

  • 期刊论文

  • 发表日期：

    2026年8月4日

查看全文
  • Yuan Zhang,
  • Shaoyi Xu,
  • Zuhao Zhu,
  • Chong Chen,
  • Hao Wang

摘要

The robotization of rapid roadway excavation equipment is a critical trend in advancing intelligent mining. However, environmental perception in excavation roadways still relies heavily on manual close-range measurements, and effective methods for autonomously acquiring workspace information during rapid excavation remain lacking. Addressing the constraints of low illumination, confined space, and limited field of view in coal mine excavation roadways, this study employs a dual-station laser synchronous scanning system integrated with excavation equipment to capture roadway environmental data and proposes an autonomous point cloud data fusion method. First, initial pose alignment is achieved by matching keypoints between the two station point clouds. Next, the optimal overlapping feature region is dynamically extracted using mutual neighborhood search combined with regional clustering weighting, thereby establishing an adaptive extraction model for this region. Finally, the dynamic feature-optimized iterative closest point (D-FOICP) algorithm is employed to converge the error and complete data fusion. Experimental results demonstrate that the proposed method achieves high-precision point cloud fusion without coordinate guidance or artificial targets, reducing the root mean square error (RMSE) by over 40% and achieving a feature retention rate (FRR) of 96%, providing effective technical support for the autonomous acquisition of operational environmental information by rapid roadway excavation equipment.