摘要
传统的基于数据关联的同时定位与建图(simultaneous localization and mapping,SLAM)方法易引起观测与目标之间的误匹配,进而导致位姿估计精度下降.结合柱状特征提取方法和随机有限集理论,提出一种基于序贯蒙特卡罗实现的车辆3D激光SLAM方法.利用M估计抽样一致性算法从分割后的点云中提取稳定的柱状特征,捕获单帧点云中的静态存活特征和新生特征;在Rao-Blackwellized-概率假设密度同时定位与建图(Rao-Blackwellizedprobability hypothesis density-simultaneous localization and mapping,RB-PHD-SLAM)框架中引入两种特征,并运用序贯蒙特卡罗方法完成车辆轨迹概率密度和地图后验强度在帧间的传递,实现对环境特征和车辆位姿的同时估计.模拟数据集和KITTI数据集试验结果显示,与经典的FastSLAM算法相比,本文算法使车辆定位精度提升44.99%,并使环境特征位置估计和环境特征数量估计的平均误差分别降低49.24%和56.22%,显著提升了SLAM的运行精度和鲁棒性,有助于保障智能汽车的运行安全.
Traditional data association-based simultaneous localization and mapping(SLAM)methods are prone to causing mismatches between observations and targets,leading to a decrease in pose estimation accuracy.This paper proposes a 3D LiDAR SLAM method for cluttered environments by combining column feature extraction method and random finite set theory based on sequential Monte Carlo implementation.Firstly,stable column features are extracted from segmented point clouds using the M-estimator sample consensus algorithm to obtain static surviving features and new features within a single frame of point cloud data.Subsequently,two types of features are introduced into the RB-PHD-SLAM(Rao-Blackwellized-probability hypothesis density-simultaneous localization and mapping)framework,and the sequential Monte Carlo method is employed to achieve inter-frame propagation of the vehicle’s trajectory probability density and the map posterior.This enables simultaneous estimation of environmental features and vehicle poses.Evaluation results based on both simulation dataset and KITTI dataset show that,compared with the classical FastSLAM algorithm,the proposed method improves the vehicle positioning accuracy by 44.99%,and reduces the average estimation error of feature location and feature number by 49.24%and 56.22%,respectively.These results indicate that the proposed method significantly improves the accuracy and robustness of SLAM,and helps to ensure safe operation of intelligent vehicles.
作者
廖光亮
傅春耘
于水恩
李健
王建文
LIAO Guangliang;FU Chunyun;YU Shuien;LI Jian;WANG Jianwen(College of Mechanical and Vehicle Engineering,Chongqing University,Chongqing 400044,China;Chongqing Changan Industry(Group)Co.,Ltd.,Chongqing 401120,China;Chongqing Changan Automobile Co.,Ltd.,Chongqing 400023,China)
出处
《湖南大学学报(自然科学版)》
北大核心
2025年第2期64-75,共12页
Journal of Hunan University:Natural Sciences
基金
重庆市技术创新与应用发展专项重点项目(cstc2021jscx-dxwtBX0023)。
关键词
智能车辆
位置测量
特征提取
随机有限集
序贯蒙特卡罗
intelligent vehicle
position measurement
feature extraction
random finite sets
sequential Monte Carlo
