【摘要】 在电动汽车电芯极柱的激光清洗领域，精确定位清洗点是确保激光清洗效果高质量的关键前提。为提高定位的准确性，提出一种基于改进区域生长的电芯极柱点云定位方法，通过自适应半径和预设路径引导点云区域生长，特别是在边缘区域引入注意力机制，基于小范围邻域和欧式距离标准差比率进行更精确的点云分割。实验结果显示，该方法X轴和Y轴的定位误差保持在±0.6 mm以内，Z轴的定位误差在±1.5 mm以内，均方根误差相比传统区域生长点云分割在3个维度上分别降低了54.32%、50.61%、11.97%。由此证明，该方法有助于提升激光清洗的定位精度与清洗效率。更多还原

【Abstract】 In the field of laser cleaning for electric vehicle battery cell electrode poles, precisely locating the cleaning points is a key prerequisite for ensuring the high quality of the laser cleaning results. To enhance the precision of localization, this paper proposes a point cloud localization method for battery cell electrode poles based on an enhanced region growth algorithm. This method guides the growth of point clouds through an adaptive radius and preset pathways, with a particular emphasis on introducing an attention mechanism for edge areas, enabling more precise point cloud segmentation using a small neighborhood range and the standard deviation ratio of Euclidean distances. Experimental results indicate that the method achieves a localization error within ±0.6 mm for the X and Y axes, and within ±1.5 mm for the Z-axis, with a reduction in the root mean square error by 54.32%, 50.61%, and 11.97% across the three dimensions, respectively, compared to traditional region growth point cloud segmentation. These findings demonstrate that the method significantly improves the accuracy and efficiency of localization for laser cleaning.更多还原