全文下载：
202003015.pdf
文章编号： 1672-6987（2020）03-0111-08； DOI： 10.16351/j.1672-6987.2020.03.015
马立修a， 李硕贤b， 翟常冬a， 王永威a
(山东理工大学 a.电气与电子工程学院;b.交通与车辆工程学院，山东 淄博 255049）
摘要： 研究了一种基于旋转激光台扫描识别圆柱标志物的小区域定位系统，并在此基础上给出理想路径的跟随控制策略，解决了在拒止环境下因定位精度低而无法满足路径跟随控制要求的问题。由旋转激光台扫描得到圆柱形标志物横截面部分轮廓的点云图，对其采用RANSAC圆弧识别算法以定位轮廓中心，通过串口将坐标数据发送至STM32单片机，再结合角度传感器测得的车头偏航角，经过直角坐标变换，得到以标志物圆心为原点的静止坐标系下的车辆位置。而后基于定位结果给出行驶路线，利用距离偏差和角度偏差对车辆转向机进行PD控制，以沿路径行驶。基于平台模型实验结果表明：本工作提出的8种车头方向情况中，定位的最大横向偏差为+4 cm，最大纵向偏差为-4 cm，达到较高的精度；RANSAC圆弧识别算法对标志物距离较远或激光雷达精度不高的场景有较好的适应性。
关键词： 拒止环境； 旋转激光台； RANSAC圆弧识别算法； 定位算法； 路径规划； 偏差控制； 惯性导航
中图分类号： TP 391.41文献标志码： A
引用格式： 马立修， 李硕贤， 翟常冬， 等. 拒止环境下的旋转激光台RANSAC识别定位算法\[J\]. 青岛科技大学学报（自然科学版）， 2020， 41（3）： 111-118.
MA Lixiu， LI Shuoxian， ZHAI Changdong, et al. Rotating laser station RANSAC recognition and location algorithm in rejected environment\[J\]. Journal of Qingdao University of Science and Technology（Natural Science Edition), 2020， 41（3）： 111-118.
Rotating Laser Station RANSAC Recognition and Location
Algorithm in Rejected EnvironmentMA Lixiua， LI Shuoxianb， ZHAI Changdonga， WANG Yongweia
(a.School of Electrical and Electronic Engineering; b.School of Transportation and Vehicle Engineering,
Shandong University of Technology, Zibo 255049， China)Abstract: A small-area positioning system based on rotating laser scanner to identify cylindrical markers is studied. On this basis, an ideal path following control strategy is given, which solves the problem that the path following control requirements can not be met due to the low positioning accuracy in the rejection environment. The point clouds of the cross-section contour of cylindrical markers are obtained by scanning with a rotating laser station. RANSAC arc search algorithm is used to locate the center of the contour. The coordinate data is sent to STM32 MCU through serial port, and then combined with the yaw angle of vehicle measured by angle sensor, the vehicle position in the stationary coordinate system with the center of the marker as the origin is obtained by rectangular coordinate transformation. Then, the driving route is given based on the positioning results, and the vehicle steering gear is controlled by PD using distance and angle deviation to drive along the path. The experimental results based on the platform model show that the maximum lateral deviation of positioning is +4 cm and the maximum longitudinal deviation is -4 cm, which achieves high accuracy. The RANSAC arc search algorithm has better adaptability to the scene where the marker is farther or the lidar accuracy is not high.
Key words: rejected environment； rotating laser station； RANSAC arc search algorithm； location algorithm； route plan； deviation control； inertial navigation
收稿日期： 2019-05-05
作者简介： 马立修（1971—），男，副教授.
