针对目前移动机器人对环境地图构建精度低的问题, 分别提出了激光雷达测距、测角的标定方法。通过误差传播定律分析引起激光雷达的测距误差因素, 可知激光雷达测距误差主要是由回波强度和测量距离引起的, 推导出测距误差修正模型。通过分析激光雷达测角误差因素, 针对机械扫描轴与几何旋转中心偏心引起的误差, 提出了一种三角形标定方法, 建立测角误差修正模型。根据激光雷达测距、测角误差修正模型修改移动机器人坐标转换系统。实验结果表明, 测距标定使平面障碍物数据纵坐标差值的标准差提高了30%~60%, 接近物体真实几何特征; 测角标定方法使障碍物数据的重合效果提高了30%, 标定方法提高了移动机器人地图构建的精度。

Aiming at the problem that the current mobile robots have low accuracy for the construction of environmental maps, the calibration methods of ranging and angle measurement of LiDAR were proposed respectively. The error propagation law was used to analyze the ranging error factor of the LiDAR. It can be seen that the LiDAR ranging error was mainly caused by the echo intensity and the measuring distance, and the ranging error correction model was derived. By analyzing the error factors of LiDAR angle measurement, a triangulation calibration method was proposed for the error caused by the eccentricity of the mechanical scanning axis and the geometric rotation center, and the angle error correction model was established. The mobile robot coordinate conversion system was modified according to the LiDAR ranging and the angle correction model. The experimental results show that the standardization of the distance measurement increases the standard deviation of the longitudinal coordinate difference of the plane obstacle data by 30%-60%, which is close to the real geometric feature of the object. The angle measurement method improves the coincidence effect of the obstacle data by 30%. The accuracy of map construction of mobile robots is improved using the calibration method.