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-Blackwellized-probability 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.
Abstract
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.
在智能驾驶领域中,同时定位与建图(simultaneous localization and mapping,SLAM)是指未知环境中的车辆利用特定传感器获取环境信息,同时创建地图模型并估计自身运动的过程[1]. 与视觉传感器相比,激光雷达不受环境光照影响,能更准确测量环境深度信息,因此激光SLAM在自动驾驶领域得到广泛应用[2-3].
HANY F, LIJ J, XIAOK .Research on mapping of shotcrete robot in roadway based on optimized Gmapping algorithm[J].Journal of Hunan University (Natural Sciences),2023,50(6):118-126.(in Chinese)
[3]
ZHENGX, ZHUJ K .Efficient LiDAR odometry for autonomous driving[J].IEEE Robotics and Automation Letters,2021,6(4):8458-8465.
QINX H, ZHOUH, LIAOY F,et al .Robust laser SLAM system based on temporal sliding window in dynamic scenes[J].Journal of Hunan University (Natural Sciences),2023,50(12):49-58.(in Chinese)
[6]
SMITHR C, CHEESEMANP. On the representation and estimation of spatial uncertainty[J]. The International Journal of Robotics Research,1986, 5(4): 56-68.
[7]
MONTEMERLOM, THRUNS, KOLLERD, et al. FastSLAM: a factored solution to the simultaneous localization and mapping problem[C]// Proceedings of the AAAI National Conference on Artificial Intelligence, July 28-August 1, 2002. Edmonton, Canada: AAAI, 2002: 592-598.
[8]
SINGERR A, SEAR G. New results in optimizing surveillance system tracking and data correlation performance in dense multitarget environments[J]. IEEE Transactions on Edmonton Control, 1973, 18(6): 571-582.
[9]
NEIRAJ, TARDOSJ D .Data association in stochastic mapping using the joint compatibility test[J].IEEE Transactions on Robotics and Automation, 2001, 17(6): 890-897.
[10]
MULLANEJ, VOB N, ADAMSM D .Rao-Blackwellised PHD SLAM[C]//2010 IEEE International Conference on Robotics and Automation, May 3-7,2010. Anchorage,AK, USA: IEEE,2010: 5410-5416.
[11]
MULLANEJ, VOB N, ADAMSM D,et al .A random-finite-set approach to Bayesian SLAM[J].IEEE Transactions on Robotics,2011, 27(2): 268-282.
[12]
GRISETTIG, STACHNISSC, BURGARDW .Improving grid-based SLAM with Rao-Blackwellized particle filters by adaptive proposals and selective resampling[C]//Proceedings of the 2005 IEEE International Conference on Robotics and Automation, April 18-22,2005. Barcelona, Spain: IEEE,2005:2432-2437.
[13]
ADAMSM, VOB N, MAHLERR,et al .SLAM gets a PHD:new concepts in map estimation[J].IEEE Robotics & Automation Magazine, 2014, 21(2): 26-37.
[14]
GAOL, BATTISTELLIG, CHISCIL .PHD-SLAM 2.0:efficient SLAM in the presence of missdetections and clutter[J].IEEE Transactions on Robotics, 2021, 37(5): 1834-1843.
[15]
VOB N, VOB T, HOANGH G .An efficient implementation of the generalized labeled multi-Bernoulli filter[J].IEEE Transactions on Signal Processing,2017,65(8):1975-1987.
[16]
VOB N, MAW K .The Gaussian mixture probability hypothesis density filter[J].IEEE Transactions on Signal Processing,2006,54(11):4091-4104.
[17]
VOB N, SINGHS, DOUCETA .Sequential Monte Carlo methods for multitarget filtering with random finite sets[J].IEEE Transactions on Aerospace and Electronic Systems,2005,41(4):1224-1245.
HANY F, TANGC C, XIAOK .RGB-D dense map construction based on improved ORB-SLAM2 algorithm[J].Journal of Hunan University (Natural Sciences),2023,50(2):52-62.(in Chinese)
[20]
DAIA, LUNDG, GAOG. PlaneSLAM:plane-based LiDAR SLAM for motion planning in structured 3D environments[EB/OL]. (2022-09-08)[2024-01-05].
[21]
CHGHAFM, RODRIGUEZS, OUARDI AEL .Camera,LiDAR and multi-modal SLAM systems for autonomous ground vehicles:a survey[J].Journal of Intelligent & Robotic Systems, 2022, 105(2): 1-35.
[22]
SONGF, SHAOY T, GAOW,et al. Layer-wise geometry aggregation framework for lossless LiDAR point cloud compression[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2021, 31(12): 4603-4616.
[23]
KONGJ, PFEIFFERM, SCHILDBACHG, et al. Kinematic and dynamic vehicle models for autonomous driving control design[C]//2015 IEEE Intelligent Vehicles Symposium (IV),June 28 -July 1,2015. Seoul,Korea: IEEE,2015:1094-1099.
[24]
RENY, LIUB B, CHENGR, et al .Lightweight semantic-aided localization with spinning LiDAR sensor[J].IEEE Transactions on Intelligent Vehicles,2023,8(1):605-615.
[25]
EBADIK, PALIERIM, WOODS,et al .DARE-SLAM:degeneracy-aware and resilient loop closing in perceptually-degraded environments[J].Journal of Intelligent & Robotic Systems,2021,102(2):1-25.
[26]
ANANDB, SENAPATIM, BARSAIYANV,et al .LiDAR-INS/GNSS-based real-time ground removal,segmentation,and georeferencing framework for smart transportation[J].IEEE Transactions on Instrumentation and Measurement,2021,70:8504611.
[27]
KRISHNAG S, SUPRIYAK, BAIDYAS .3DS-SLAM:a 3D object detection based semantic SLAM towards dynamic indoor environments[EB/OL]. (2023-10-10)[2024-01-05].
[28]
MULLANEJ, VOB N, ADAMSM D,et al .A random set formulation for Bayesian SLAM[C]//2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, September 22-26,2008. Nice, France: IEEE,2008:1043-1049.
[29]
ZHUY, LIANGS, LIP,et al .Robust human pose quality assessment using optimal sub-pattern assignment[C]//2022 11th International Conference on Control,Automation and Information Sciences (ICCAIS),November 21-24,2022. Hanoi,Vietnam:IEEE, 2022: 419-423.