同时定位与建图(simultaneous localization and mapping,SLAM)是地下空间自主探测、自动巡检和应急救援的关键。然而,地下空间巷道狭长、地形复杂、光照不均等使得激光点云和视觉图像匹配极易发生退化,进而导致多源传感器数据融合SLAM...同时定位与建图(simultaneous localization and mapping,SLAM)是地下空间自主探测、自动巡检和应急救援的关键。然而,地下空间巷道狭长、地形复杂、光照不均等使得激光点云和视觉图像匹配极易发生退化,进而导致多源传感器数据融合SLAM精度不足,甚至失效。为此,本文提出一种增强稳健性的多源传感器数据动态加权融合SLAM方法。首先,在视觉图像预处理阶段,采用了一种基于色调、饱和度、亮度(hue,stauration,value,HSV)空间的图像增强技术,结合单参数同态滤波和对比度受限的自适应直方图均衡化(contrast limited adaptive histogram equalization,CLAHE)算法,有效提升了地下空间图像的亮度和对比度,从而增强了视觉里程计的稳健性。然后,通过马氏距离一致性检验方法对各传感器的数据质量进行评估,分析数据退化情况,并自适应地选择适合当前场景的传感器数据进行融合。最后,在综合考虑各传感器关键参数的基础上,构建了多源传感器因子图模型,并根据数据质量动态调整各传感器数据融合因子的权重,形成多源传感器数据权重动态组合模型。为验证本文方法的有效性,使用自主设计集成的移动机器人在地下走廊、开挖的地铁隧道和煤矿巷道等典型地下空间中分别进行了试验,并与多种主流SLAM方法进行定性、定量对比分析。结果表明:本文方法最大轨迹均方根误差(root mean square error,RMSE)仅为0.19 m,以高精度地面三维激光扫描获取的点云为参考,平均点云直接距离比较(cloud to cloud,C2C)小于0.13 m,所构建的点云地图具有较好的全局一致性和几何结构真实性,验证了本文方法在复杂地下空间具有更高的精度和稳健性。展开更多
高速公路下穿涵洞积水经常带来交通堵塞,人员伤亡和财产损失等问题,为了解决此类问题,设计出一款基于双波长数据融合的高速公路涵洞积水便携式无线传感器。该传感器由锂电池、毫米波雷达、激光雷达、4G通讯模块和控制电路等部分组成,通...高速公路下穿涵洞积水经常带来交通堵塞,人员伤亡和财产损失等问题,为了解决此类问题,设计出一款基于双波长数据融合的高速公路涵洞积水便携式无线传感器。该传感器由锂电池、毫米波雷达、激光雷达、4G通讯模块和控制电路等部分组成,通过算法设计实现毫米波雷达数据和激光雷达数据的双波长融合。该传感器在杭州绕城西复线S43湖州段内5处下穿涵洞及杭宁高速2处涵洞完成了安装布设,实测传感精度优于± 3 mm。本文设计实现的基于双波长数据融合的便携式无线传感器,可应用于高速公路涵洞积水监测,具有推广应用价值。Accumulation of water in highway culverts often leads to traffic congestion, casualties, and property damage. To address these issues, a portable wireless sensor for highway culvert water accumulation based on dual-wavelength data fusion has been designed. The sensor mainly consists of a lithium battery, millimeter-wave radar, lidar, 4G communication module, and control circuit. It achieves dual-wavelength fusion of millimeter-wave radar data and lidar data through algorithm design. The sensor has been installed in five underpasses along the Hangzhou Western Expressway S43 Huzhou section, with a measured sensing distance of up to 10 meters and a sensing accuracy better than 3mm. The portable wireless sensor designed and implemented in this paper, based on dual-wavelength data fusion, can be applied to the monitoring of water accumulation in highway culverts and has the value of promotion and application.展开更多
文摘同时定位与建图(simultaneous localization and mapping,SLAM)是地下空间自主探测、自动巡检和应急救援的关键。然而,地下空间巷道狭长、地形复杂、光照不均等使得激光点云和视觉图像匹配极易发生退化,进而导致多源传感器数据融合SLAM精度不足,甚至失效。为此,本文提出一种增强稳健性的多源传感器数据动态加权融合SLAM方法。首先,在视觉图像预处理阶段,采用了一种基于色调、饱和度、亮度(hue,stauration,value,HSV)空间的图像增强技术,结合单参数同态滤波和对比度受限的自适应直方图均衡化(contrast limited adaptive histogram equalization,CLAHE)算法,有效提升了地下空间图像的亮度和对比度,从而增强了视觉里程计的稳健性。然后,通过马氏距离一致性检验方法对各传感器的数据质量进行评估,分析数据退化情况,并自适应地选择适合当前场景的传感器数据进行融合。最后,在综合考虑各传感器关键参数的基础上,构建了多源传感器因子图模型,并根据数据质量动态调整各传感器数据融合因子的权重,形成多源传感器数据权重动态组合模型。为验证本文方法的有效性,使用自主设计集成的移动机器人在地下走廊、开挖的地铁隧道和煤矿巷道等典型地下空间中分别进行了试验,并与多种主流SLAM方法进行定性、定量对比分析。结果表明:本文方法最大轨迹均方根误差(root mean square error,RMSE)仅为0.19 m,以高精度地面三维激光扫描获取的点云为参考,平均点云直接距离比较(cloud to cloud,C2C)小于0.13 m,所构建的点云地图具有较好的全局一致性和几何结构真实性,验证了本文方法在复杂地下空间具有更高的精度和稳健性。
文摘高速公路下穿涵洞积水经常带来交通堵塞,人员伤亡和财产损失等问题,为了解决此类问题,设计出一款基于双波长数据融合的高速公路涵洞积水便携式无线传感器。该传感器由锂电池、毫米波雷达、激光雷达、4G通讯模块和控制电路等部分组成,通过算法设计实现毫米波雷达数据和激光雷达数据的双波长融合。该传感器在杭州绕城西复线S43湖州段内5处下穿涵洞及杭宁高速2处涵洞完成了安装布设,实测传感精度优于± 3 mm。本文设计实现的基于双波长数据融合的便携式无线传感器,可应用于高速公路涵洞积水监测,具有推广应用价值。Accumulation of water in highway culverts often leads to traffic congestion, casualties, and property damage. To address these issues, a portable wireless sensor for highway culvert water accumulation based on dual-wavelength data fusion has been designed. The sensor mainly consists of a lithium battery, millimeter-wave radar, lidar, 4G communication module, and control circuit. It achieves dual-wavelength fusion of millimeter-wave radar data and lidar data through algorithm design. The sensor has been installed in five underpasses along the Hangzhou Western Expressway S43 Huzhou section, with a measured sensing distance of up to 10 meters and a sensing accuracy better than 3mm. The portable wireless sensor designed and implemented in this paper, based on dual-wavelength data fusion, can be applied to the monitoring of water accumulation in highway culverts and has the value of promotion and application.
