摘要
为了准确获取机器人周围的环境信息,将单线激光雷达应用在地面车辆构建的地面环境感知实验系统。利用阿基米德螺线凸轮机构实现UTM-LX30单线激光雷达的匀速往复运动,通过数据拼接算法和空间几何变换实现场景重建和障碍特征提取,通过插值拟合成统一格式的面数据;将雷达扫描所得地形图中高点和低点进行匹配,计算得到最大偏差为0.5 cm,完全能够满足无人平台的导航需求;运用Matlab模拟构建10 m×10 m区域,根据障碍特征提取得到3维地形数据的最大高度差和最大斜度,结合车体的最大越阶高度和最大爬坡角度,仿真得出无人平台的最佳路径。仿真结果表明:地面环境感知技术具有可行性和有效性,有一定参考价值和实用价值。
In order to obtain the information about the environment around the robot,a single laser radar is applied tothe ground environment sensing experiment system for the vehicles.Reciprocating motion at a constant speed of the singlelaser radar(UTM-LX30)is realized by using an Archimedes spiral CAM mechanism.Environment construction andobstacle identification are obtained by data stitching algorithm and space geometry transformation.The data is transformedto surface data with uniform format by interpolating fitting.Through matching the peak points and valley points on thesurface data obtained by radar scan,the maximal calculated theoretical deviation is0.5cm,which is enough to meet thenavigation requirements of the unmanned platform.10m×10m areas are built by Matlab simulation.The maximal heightdifferences and maximal inclination of the three-dimensional morphology data are obtained according to the obstacleidentification.Based on the maximal surmountable obstacle height and the maximal climbing angle of the vehicle,theoptimal path for the unmanned platform can be obtained by simulation.Result of the simulation shows that the groundenvironment sensing technology is feasible and effective,and it has a certain reference value and practical value.
作者
许刘林
沈新民
周建钊
彭康
刘晴
Xu Liulin;Shen Xinmin;Zhou Jianzhao;Peng Kang;Liu Qing(College of Field Engineering, PLA University of Science & Technology, Nanjing 210007, China)
出处
《兵工自动化》
2017年第9期17-22,49,共7页
Ordnance Industry Automation
基金
国家重点研发计划(2016YFC0802903)
国家自然科学基金青年基金项目(51505498)
江苏省自然科学基金青年基金项目(BK20150714)
关键词
无人平台
单线激光雷达
数据拼接算法
环境感知技术
障碍特征提取
路径规划
unmanned platform
single laser radar
data stitching algorithm
environment perception technology
obstacle identification
path planning
