基于激光雷达信息的履带式自动割草机障碍物检测
|
摘要
针对果园土壤生草制管理中所面临的割草任务重、劳动强度大等问题,设计一种基于二维激光雷达信息的履带式自动割草机障碍物检测方法。采用RPLIDAR A1进行障碍物扫描检测,由Matlab编程完成数据处理。将采集到的数据转换为直角坐标系下的x、y坐标,运用最小二乘滤波对数据进行去噪处理,采用k-means聚类算法对扫描数据进行聚类分析,得到障碍物的基本轮廓信息,最后经拟合得到较为密集的障碍物边缘离散点。试验结果中激光雷达扫描得到的角度和距离数据与现场实地测量数据之间最大误差为3.2%,达到了障碍物检测的目的,表明基于激光雷达信息的障碍物检测方法是合理可行的。
Aiming at the problems of heavy mowing task and high labor intensity in the orchard soil grass management, a crawler automatic mower obstacle detection method based on two-dimensional lidar information was designed. Obstacle scanning detection was performed using RPLIDAR A1 laser radar, and data processing was completed by Matlab programming. The collected data was converted into x and y coordinates in a Cartesian coordinate system, and the data was denoised by least squares filtering. The k-means clustering algorithm was used to cluster the scanned data, and the basic contour information of the obstacle was obtained. Finally, the dense points of the edge of the obstacle were obtained by fitting. In the experimental results, the maximum error between the angle and distance data obtained by laser radar scanning and the field measurement data was 3.2%, which achieves the purpose of obstacle detection. It is indicated that the obstacle detection method based on lidar information is reasonable and feasible.
Aiming at the problems of heavy mowing task and high labor intensity in the orchard soil grass management, a crawler automatic mower obstacle detection method based on two-dimensional lidar information was designed. Obstacle scanning detection was performed using RPLIDAR A1 laser radar, and data processing was completed by Matlab programming. The collected data was converted into x and y coordinates in a Cartesian coordinate system, and the data was denoised by least squares filtering. The k-means clustering algorithm was used to cluster the scanned data, and the basic contour information of the obstacle was obtained. Finally, the dense points of the edge of the obstacle were obtained by fitting. In the experimental results, the maximum error between the angle and distance data obtained by laser radar scanning and the field measurement data was 3.2%, which achieves the purpose of obstacle detection. It is indicated that the obstacle detection method based on lidar information is reasonable and feasible.
引文
[1] Almasri M,Elleithy K,Alajlan A.Sensor fusion based model for collision free mobile robot navigation [J].Sensors,2016,16(1):1-24.
[2] Keller C G,Enzweiler M,Rohrbach M,et al.The benefits of dense stereo for pedestrian detection [J].IEEE Transactions on Intelligent Transportation Systems,2011,12(4):1096-1106.
[3] 房宏,于晶晶.基于激光传感器的机器人位姿跟踪[J].激光杂志,2015,36(10):91-94.Fang Hong,Yu Jingjing.Robots pose tracking based on laser sensor [J].Laser Journal,2015,36(10):91-94.
[4] 贺红林,何文丛,刘文光,等.神经网络与计算力矩复合的机器人运动轨迹跟踪控制[J].农业机械学报,2013,44(5):270-275.He Honglin,He Wencong,Liu Wenguang,et al.Tracking control of robot using hybrid controller based on neural network and computed torque [J].Transactions of the Chinese Society of Agricultural Machinery,2013,44(5):270-275.
[5] 柯文德,彭志平,蔡则苏,等.仿人机器人相似性运动轨迹跟踪控制研究[J].自动化学报,2014,40(11):2404-2413.Ke Wende,Peng Zhiping,Cai Zesu,et al.Study of trajectory tracking control for humanoid robot based on similarity locomotion [J].Acta Automatica Sinica,2014,40(11):2404-2413.
[6] 王会,侯勇,侯世亮.一种基于激光雷达无人车的障碍物探测系统设计[J].北京工业职业技术学院学报,2018,17(1):4-7,22.Wang Hui,Hou Yong,Hou Shiliang.Design of laser radar obstacle detection system for unmanned vehicles [J].Journal of Beijing Institute of Industry Technology,2018,17(1):4-7,22.
[7] 沈娣丽,孟雅俊,路程,等.基于Wi-Fi的温室大棚监测控制系统[J].中国农机化,2012(1):162-165,139.Shen Dili,Meng Yajun,Lu Cheng,et al.Monitoring and control system of greenhouse based on Wi-Fi [J].Chinese Agricultural Mechanization,2012(1):162-165,139.
[8] 杨海波,杨新春,李金热,等.基于PLC的双轴工作台控制与实现[J].中国农机化学报,2014,35(4):228-231.Yang Haibo,Yang Xinchun,Li Jinre,et al.Control and realization of biaxial work bench based on PLC [J].Journal of Chinese Agricultural Mechanization,2014,35(4):228-231.
[9] 赵建伟,张宏静,王洪燕,等.应用激光雷达构建室内环境地图的研究[J].机械设计与制造,2017(5):135-137.Zhao Jianwei,Zhang Hongjing,Wang Hongyan,et al.Study on building indoor environment map with laser radar [J].Machinery Design & Manufacture,2017(5):135-137.
[10] 李健华.基于GPRS技术的膜下滴灌自动化控制系统的设计[J].中国农机化学报,2014,35(1):245-248.Li Jianhua.Automatic drip irrigation under film control system design based on GPRS technology [J].Journal of Chinese Agricultural Mechanization,2014,35(1):245-248.
[11] 王媚,朱兴林.基于车载激光雷达设备的跟驰数据采集方法[J].中国科技论文,2017,12(19):2176-2180.Wang Mei,Zhu Xinglin.Data acquisition method of car-following model for uban expressway [J].Chinese Journal of Scientific and Technical Papers,2017,12(19):2176-2180.
[12] 王靠省,郭平.地方独立坐标转换WGS84大地坐标方法及程序设计[J].山西建筑,2017,43(1):215-217.Wang Kaosheng,Guo Ping.Local independent coordinate transformation WGS84 geodetic coordinate method and program design [J].Shanxi Architecture,2017,43(1):215-217.
[13] 杨森森.基于GPS/INS/激光雷达的无人车组合导航[D].上海:上海交通大学,2013.Yang Sensen.Integrated navigation of unmanned vehicles based on GPS/INS/Lidar [D].Shanghai:Shanghai Jiaotong University,2013.
[14] 戚尔江,彭道刚,关欣蕾,等.基于RTK北斗和激光雷达的巡检机器人导航系统研究[J].仪表技术与传感器,2018(6):58-63.Qi Erjiang,Peng Daogang,Guan Xinlei,et al.Research on patrol robot navigation system based on RTK Beidou and laser radar [J].Instrument Technique and Sensor,2018(6):58-63.
[15] stu-zhang.K-Means聚类算法原理及其python和matlab实现[EB/OL] .https://blog.csdn.net/zpp1994/article/details/ 53456306,2018-11-19.
[16] 彭志,李传荣,周梅.基于最小二乘曲线拟合法的点云滤波算法研究[J].遥感信息,2011(5):86-89,113.Peng Zhi,Li Chuanrong,Zhou Mei.Study on filtering of point cloud based on least squares curve fitting method [J].Remote Sensing Information,2011(5):86-89,113.
[2] Keller C G,Enzweiler M,Rohrbach M,et al.The benefits of dense stereo for pedestrian detection [J].IEEE Transactions on Intelligent Transportation Systems,2011,12(4):1096-1106.
[3] 房宏,于晶晶.基于激光传感器的机器人位姿跟踪[J].激光杂志,2015,36(10):91-94.Fang Hong,Yu Jingjing.Robots pose tracking based on laser sensor [J].Laser Journal,2015,36(10):91-94.
[4] 贺红林,何文丛,刘文光,等.神经网络与计算力矩复合的机器人运动轨迹跟踪控制[J].农业机械学报,2013,44(5):270-275.He Honglin,He Wencong,Liu Wenguang,et al.Tracking control of robot using hybrid controller based on neural network and computed torque [J].Transactions of the Chinese Society of Agricultural Machinery,2013,44(5):270-275.
[5] 柯文德,彭志平,蔡则苏,等.仿人机器人相似性运动轨迹跟踪控制研究[J].自动化学报,2014,40(11):2404-2413.Ke Wende,Peng Zhiping,Cai Zesu,et al.Study of trajectory tracking control for humanoid robot based on similarity locomotion [J].Acta Automatica Sinica,2014,40(11):2404-2413.
[6] 王会,侯勇,侯世亮.一种基于激光雷达无人车的障碍物探测系统设计[J].北京工业职业技术学院学报,2018,17(1):4-7,22.Wang Hui,Hou Yong,Hou Shiliang.Design of laser radar obstacle detection system for unmanned vehicles [J].Journal of Beijing Institute of Industry Technology,2018,17(1):4-7,22.
[7] 沈娣丽,孟雅俊,路程,等.基于Wi-Fi的温室大棚监测控制系统[J].中国农机化,2012(1):162-165,139.Shen Dili,Meng Yajun,Lu Cheng,et al.Monitoring and control system of greenhouse based on Wi-Fi [J].Chinese Agricultural Mechanization,2012(1):162-165,139.
[8] 杨海波,杨新春,李金热,等.基于PLC的双轴工作台控制与实现[J].中国农机化学报,2014,35(4):228-231.Yang Haibo,Yang Xinchun,Li Jinre,et al.Control and realization of biaxial work bench based on PLC [J].Journal of Chinese Agricultural Mechanization,2014,35(4):228-231.
[9] 赵建伟,张宏静,王洪燕,等.应用激光雷达构建室内环境地图的研究[J].机械设计与制造,2017(5):135-137.Zhao Jianwei,Zhang Hongjing,Wang Hongyan,et al.Study on building indoor environment map with laser radar [J].Machinery Design & Manufacture,2017(5):135-137.
[10] 李健华.基于GPRS技术的膜下滴灌自动化控制系统的设计[J].中国农机化学报,2014,35(1):245-248.Li Jianhua.Automatic drip irrigation under film control system design based on GPRS technology [J].Journal of Chinese Agricultural Mechanization,2014,35(1):245-248.
[11] 王媚,朱兴林.基于车载激光雷达设备的跟驰数据采集方法[J].中国科技论文,2017,12(19):2176-2180.Wang Mei,Zhu Xinglin.Data acquisition method of car-following model for uban expressway [J].Chinese Journal of Scientific and Technical Papers,2017,12(19):2176-2180.
[12] 王靠省,郭平.地方独立坐标转换WGS84大地坐标方法及程序设计[J].山西建筑,2017,43(1):215-217.Wang Kaosheng,Guo Ping.Local independent coordinate transformation WGS84 geodetic coordinate method and program design [J].Shanxi Architecture,2017,43(1):215-217.
[13] 杨森森.基于GPS/INS/激光雷达的无人车组合导航[D].上海:上海交通大学,2013.Yang Sensen.Integrated navigation of unmanned vehicles based on GPS/INS/Lidar [D].Shanghai:Shanghai Jiaotong University,2013.
[14] 戚尔江,彭道刚,关欣蕾,等.基于RTK北斗和激光雷达的巡检机器人导航系统研究[J].仪表技术与传感器,2018(6):58-63.Qi Erjiang,Peng Daogang,Guan Xinlei,et al.Research on patrol robot navigation system based on RTK Beidou and laser radar [J].Instrument Technique and Sensor,2018(6):58-63.
[15] stu-zhang.K-Means聚类算法原理及其python和matlab实现[EB/OL] .https://blog.csdn.net/zpp1994/article/details/ 53456306,2018-11-19.
[16] 彭志,李传荣,周梅.基于最小二乘曲线拟合法的点云滤波算法研究[J].遥感信息,2011(5):86-89,113.Peng Zhi,Li Chuanrong,Zhou Mei.Study on filtering of point cloud based on least squares curve fitting method [J].Remote Sensing Information,2011(5):86-89,113.