An increasing number of drivers are relying on digital map navigation systems in vehicles or mobile phones to select optimal driving routes in order to save time and improve safety. In the near future, digital map nav...An increasing number of drivers are relying on digital map navigation systems in vehicles or mobile phones to select optimal driving routes in order to save time and improve safety. In the near future, digital map navigation systems are expected to play more important roles in transportation systems. In order to extend current navigation systems to more applications, two fundamental problems must be resolved: the lane-level map model and lane-level route planning. This study proposes solutions to both problems. The current limitation of the lane-level map model is not its accuracy but its flexibility;this study proposes a novel seven-layer map structure, called as Tsinghua map model, which is able to support autonomous driving in a flexible and efficient way. For lane-level route planning, we propose a hierarchical route-searching algorithm to accelerate the planning process, even in the presence of complicated lane networks. In addition, we model the travel costs allocated for lane-level road networks by analyzing vehicle maneuvers in traversing lanes, changing lanes, and turning at intersections. Tests were performed on both a grid network and a real lane-level road network to demonstrate the validity and efficiency of the proposed algorithm.展开更多
A lane-level intersection map is a cornerstone in high-definition(HD) traffic network maps for autonomous driving and high-precision intelligent transportation systems applications such as traffic management and contr...A lane-level intersection map is a cornerstone in high-definition(HD) traffic network maps for autonomous driving and high-precision intelligent transportation systems applications such as traffic management and control, and traffic accident evaluation and prevention. Mapping an HD intersection is time-consuming, labor-intensive, and expensive with conventional methods. In this paper, we used a low-channel roadside light detection and range sensor(LiDAR) to automatically and dynamically generate a lane-level intersection, including the signal phases, geometry, layout, and lane directions. First, a mathematical model was proposed to describe the topology and detail of a lane-level intersection. Second, continuous and discontinuous traffic object trajectories were extracted to identify the signal phases and times. Third, the layout, geometry, and lane direction were identified using the convex hull detection algorithm for trajectories. Fourth, a sliding window algorithm was presented to detect the lane marking and extract the lane, and the virtual lane connecting the inbound and outbound of the intersection were generated using the vehicle trajectories within the intersection and considering the traffic rules. In the field experiment, the mean absolute estimation error is 2 s for signal phase and time identification. The lane marking identification Precision and Recall are96% and 94.12%, respectively. Compared with the satellite-based,MMS-based, and crowdsourcing-based lane mapping methods,the average lane location deviation is 0.2 m and the update period is less than one hour by the proposed method with low-channel roadside LiDAR.展开更多
The essential requirement for precise localization of a self-driving car is a lane-level map which includes road markings(RMs).Obviously,we can build the lane-level map by running a mobile mapping system(MMS)which is ...The essential requirement for precise localization of a self-driving car is a lane-level map which includes road markings(RMs).Obviously,we can build the lane-level map by running a mobile mapping system(MMS)which is equipped with a high-end 3D LiDAR and a number of high-cost sensors.This approach,however,is highly expensive and ineffective since a single high-end MMS must visit every place for mapping.In this paper,a lane-level RM mapping system using a monocular camera is developed.The developed system can be considered as an alternative to expensive high-end MMS.The developed RM map includes the information of road lanes(RLs)and symbolic road markings(SRMs).First,to build a lane-level RM map,the RMs are segmented at pixel level through the deep learning network.The network is named RMNet.The segmented RMs are then gathered to build a lane-level RM map.Second,the lane-level map is improved through loop-closure detection and graph optimization.To train the RMNet and build a lane-level RM map,a new dataset named SeRM set is developed.The set is a large dataset for lane-level RM mapping and it includes a total of 25157 pixel-wise annotated images and 21000 position labeled images.Finally,the proposed lane-level map building method is applied to SeRM set and its validity is demonstrated through experimentation.展开更多
精确的环境感知是实现自主代客泊车(automated valet parking,AVP)功能的基础,传统的AVP系统主要依赖于单车的感知,但随着场端智能技术的不断发展,车端与场端之间协同交互成为自主代客泊车落地的必然趋势。本文提出了一种基于V2X车场协...精确的环境感知是实现自主代客泊车(automated valet parking,AVP)功能的基础,传统的AVP系统主要依赖于单车的感知,但随着场端智能技术的不断发展,车端与场端之间协同交互成为自主代客泊车落地的必然趋势。本文提出了一种基于V2X车场协同的地下停车场全域感知方法,该方法将地下停车场的全域感知问题转化为大规模图模型的构建与优化问题。通过输入场端激光雷达、摄像头的传感器信息以及智能网联车的感知数据,以车辆位姿为节点,建立多种边约束关系。为了提高感知精度,本文提出了一种融合车道级地图信息的大规模图模型方法,通过将停放车辆作为半静态信息约束,并结合车道级地图信息构建横向约束,在求解过程中引入滑动窗口以减小图模型的规模,最终以地图形式输出感知结果供车端使用。通过仿真实验和在占地面积为2 500 m^(2)以上的地下停车场场景中进行实地实验,结果表明,该方法显著提升了在复杂停车场环境下的感知能力,实现了地下停车场的全域感知。展开更多
为提升自动驾驶车辆在多车道行驶与作业时的道路环境感知能力,提出了自动驾驶环境下车道级雷视融合方法 LLV-SLAM(lane-level LiDAR-visual fusion SLAM),并构建了适用于雷视融合的实时定位与建图算法(simultaneous localization and ma...为提升自动驾驶车辆在多车道行驶与作业时的道路环境感知能力,提出了自动驾驶环境下车道级雷视融合方法 LLV-SLAM(lane-level LiDAR-visual fusion SLAM),并构建了适用于雷视融合的实时定位与建图算法(simultaneous localization and mapping,SLAM)。首先,在视觉特征点提取的基础上引入直方图均衡化,并利用激光雷达获取特征点深度信息,通过视觉特征跟踪以提升SLAM系统鲁棒性。其次,利用视觉关键帧信息对激光点云进行运动畸变校正,并将LeGO-LOAM(lightweight and groud-optimized lidar odometry and mapping)融入视觉ORBSLAM2(oriented FAST and rotated BRIEF SLAM2)以增强闭环检测与矫正性能,降低系统累计误差。最后,将视觉图像所获取的位姿进行坐标转换作为激光里程计的位姿初值,辅助激光雷达SLAM进行三维场景重建。实验结果表明:相比于传统的SLAM方法,融合后的LLV-SLAM方法平均定位时延减少了41.61%;在x、y、z方向上的平均定位误差分别减少了34.63%、38.16%、24.09%;在滚转角、俯仰角、偏航角方向上的平均旋转误差减少了40.8%、37.52%、39.5%。LLV-SLAM算法有效抑制了LeGO-LOAM算法的尺度漂移,实时性和鲁棒性有显著提升,能够满足自动驾驶车辆对多车道道路环境的感知需要。展开更多
基金the National Key Research and Development Program of China (2018YFB0105000)the National Natural Science Foundation of China (61773234 and U1864203)+2 种基金the Project of Tsinghua University and Toyota Joint Research Center for AI Technology of Automated Vehicle (TT2018-02)the International Science and Technology Cooperation Program of China (2016YFE0102200)the software developed in the Beijing Municipal Science and Technology Program (D171100005117001 and Z181100005918001).
文摘An increasing number of drivers are relying on digital map navigation systems in vehicles or mobile phones to select optimal driving routes in order to save time and improve safety. In the near future, digital map navigation systems are expected to play more important roles in transportation systems. In order to extend current navigation systems to more applications, two fundamental problems must be resolved: the lane-level map model and lane-level route planning. This study proposes solutions to both problems. The current limitation of the lane-level map model is not its accuracy but its flexibility;this study proposes a novel seven-layer map structure, called as Tsinghua map model, which is able to support autonomous driving in a flexible and efficient way. For lane-level route planning, we propose a hierarchical route-searching algorithm to accelerate the planning process, even in the presence of complicated lane networks. In addition, we model the travel costs allocated for lane-level road networks by analyzing vehicle maneuvers in traversing lanes, changing lanes, and turning at intersections. Tests were performed on both a grid network and a real lane-level road network to demonstrate the validity and efficiency of the proposed algorithm.
基金supported in part by the Scientific Research Project of the Education Department of Jilin Province (JJKH20221020KJ)the National Natural Science Foundation of China (51408257)the Graduate Innovation Fund of Jilin University (101832020CX150)。
文摘A lane-level intersection map is a cornerstone in high-definition(HD) traffic network maps for autonomous driving and high-precision intelligent transportation systems applications such as traffic management and control, and traffic accident evaluation and prevention. Mapping an HD intersection is time-consuming, labor-intensive, and expensive with conventional methods. In this paper, we used a low-channel roadside light detection and range sensor(LiDAR) to automatically and dynamically generate a lane-level intersection, including the signal phases, geometry, layout, and lane directions. First, a mathematical model was proposed to describe the topology and detail of a lane-level intersection. Second, continuous and discontinuous traffic object trajectories were extracted to identify the signal phases and times. Third, the layout, geometry, and lane direction were identified using the convex hull detection algorithm for trajectories. Fourth, a sliding window algorithm was presented to detect the lane marking and extract the lane, and the virtual lane connecting the inbound and outbound of the intersection were generated using the vehicle trajectories within the intersection and considering the traffic rules. In the field experiment, the mean absolute estimation error is 2 s for signal phase and time identification. The lane marking identification Precision and Recall are96% and 94.12%, respectively. Compared with the satellite-based,MMS-based, and crowdsourcing-based lane mapping methods,the average lane location deviation is 0.2 m and the update period is less than one hour by the proposed method with low-channel roadside LiDAR.
基金This work was supported by the Industry Core Technology Development Project,20005062Development of Artificial Intelligence Robot Autonomous Navigation Technology for Agile Movement in Crowded Space,funded by the Ministry of Trade,industry&Energy(MOTIE,Republic of Korea).
文摘The essential requirement for precise localization of a self-driving car is a lane-level map which includes road markings(RMs).Obviously,we can build the lane-level map by running a mobile mapping system(MMS)which is equipped with a high-end 3D LiDAR and a number of high-cost sensors.This approach,however,is highly expensive and ineffective since a single high-end MMS must visit every place for mapping.In this paper,a lane-level RM mapping system using a monocular camera is developed.The developed system can be considered as an alternative to expensive high-end MMS.The developed RM map includes the information of road lanes(RLs)and symbolic road markings(SRMs).First,to build a lane-level RM map,the RMs are segmented at pixel level through the deep learning network.The network is named RMNet.The segmented RMs are then gathered to build a lane-level RM map.Second,the lane-level map is improved through loop-closure detection and graph optimization.To train the RMNet and build a lane-level RM map,a new dataset named SeRM set is developed.The set is a large dataset for lane-level RM mapping and it includes a total of 25157 pixel-wise annotated images and 21000 position labeled images.Finally,the proposed lane-level map building method is applied to SeRM set and its validity is demonstrated through experimentation.
文摘精确的环境感知是实现自主代客泊车(automated valet parking,AVP)功能的基础,传统的AVP系统主要依赖于单车的感知,但随着场端智能技术的不断发展,车端与场端之间协同交互成为自主代客泊车落地的必然趋势。本文提出了一种基于V2X车场协同的地下停车场全域感知方法,该方法将地下停车场的全域感知问题转化为大规模图模型的构建与优化问题。通过输入场端激光雷达、摄像头的传感器信息以及智能网联车的感知数据,以车辆位姿为节点,建立多种边约束关系。为了提高感知精度,本文提出了一种融合车道级地图信息的大规模图模型方法,通过将停放车辆作为半静态信息约束,并结合车道级地图信息构建横向约束,在求解过程中引入滑动窗口以减小图模型的规模,最终以地图形式输出感知结果供车端使用。通过仿真实验和在占地面积为2 500 m^(2)以上的地下停车场场景中进行实地实验,结果表明,该方法显著提升了在复杂停车场环境下的感知能力,实现了地下停车场的全域感知。
文摘为提升自动驾驶车辆在多车道行驶与作业时的道路环境感知能力,提出了自动驾驶环境下车道级雷视融合方法 LLV-SLAM(lane-level LiDAR-visual fusion SLAM),并构建了适用于雷视融合的实时定位与建图算法(simultaneous localization and mapping,SLAM)。首先,在视觉特征点提取的基础上引入直方图均衡化,并利用激光雷达获取特征点深度信息,通过视觉特征跟踪以提升SLAM系统鲁棒性。其次,利用视觉关键帧信息对激光点云进行运动畸变校正,并将LeGO-LOAM(lightweight and groud-optimized lidar odometry and mapping)融入视觉ORBSLAM2(oriented FAST and rotated BRIEF SLAM2)以增强闭环检测与矫正性能,降低系统累计误差。最后,将视觉图像所获取的位姿进行坐标转换作为激光里程计的位姿初值,辅助激光雷达SLAM进行三维场景重建。实验结果表明:相比于传统的SLAM方法,融合后的LLV-SLAM方法平均定位时延减少了41.61%;在x、y、z方向上的平均定位误差分别减少了34.63%、38.16%、24.09%;在滚转角、俯仰角、偏航角方向上的平均旋转误差减少了40.8%、37.52%、39.5%。LLV-SLAM算法有效抑制了LeGO-LOAM算法的尺度漂移,实时性和鲁棒性有显著提升,能够满足自动驾驶车辆对多车道道路环境的感知需要。
