针对目前激光雷达数据分割算法不能适应环境特征确定连续准确阈值的问题, 提出一种环境特征自适应激光雷达数据分割算法。依据二维激光雷达的数据特点以及室内环境的几何特征, 以激光雷达数据的邻近点拟合虚拟环境线,以虚拟环境线和邻近激光扫描射线的交点作为参考点, 确定自适应阈值, 完成激光雷达数据的预分割。针对用上述方法完成的数据预分割结果中存在的缺陷, 提出数据预分割后伪断点的判断方法, 对算法进行了优化。并将此算法与分段阈值分割算法、线性方程阈值分割算法进行比较和分析。环境特征自适应激光雷达数据分割算法对实验数据的分割成功率达到98%, 具有更强的环境适应能力和更高的分割准确度。

In order to solve the problem that LIDAR data segmentation algorithm cannot adapt to the environmental characteristics and determine the threshold continuously and accurately, an adaptive LIDAR data segmentation algorithm based on environmental features was proposed. According to the data characteristics of the two-dimensional lidar and the geometric characteristics of the indoor environment, the virtual environment line was fitted with the adjacent point of the laser radar data. The intersection of the virtual environment line and the adjacent laser scanning ray was taken as the reference point to determine the adaptive threshold pre-segmentation of radar data. In view of the defects in the data pre segmentation results completed by the above method, a method for judging pseudo breakpoints after data pre segmentation was proposed, and the algorithm was optimized. The algorithm was compared and analyzed with piecewise threshold segmentation algorithm and linear equation threshold segmentation algorithm. The LIDAR data segmentation algorithm adapting to environmental characteristics achieves a successful segmentation rate of 98% for the experimental data, and has better environment adaptability and higher segmentation accuracy.