资源

以下是一些常用的优秀内容整理,包含数据集、开源代码、公开课等。如果大家有好的资源也欢迎在评论里分享:)

1 数据集

1.1 VIO

1.2 无人驾驶

2 开源

2.1 视觉 SLAM/VIO

名称标签简介论文地址
SVO 2.0VIOSemi-direct Visual Odometry (SVO) developed at RPGPDF
Page
https://github.com/uzh-rpg/rpg_svo_pro_open
OpenVINSVIO基于 MSCKF 实现的一整套 VIO 库PDF
Video
https://github.com/rpng/open_vins
ORB SLAM3VIO基于特征点法 ORB SLAM,包含单目、双目VIO、多地图等实现PDFhttps://github.com/UZ-SLAMLab/ORB_SLAM3
BasaltVIOBasalt: Visual-Inertial Mapping with Non-Linear Factor RecoveryPDFhttps://gitlab.com/VladyslavUsenko/basalt
ROVIOVIOROVIO (Robust Visual Inertial Odometry)
使用IEKF将视觉信息和IMU信息进行紧耦合的VIO系统
PDF
Video
https://github.com/ethz-asl/rovio
VINS FusionVIO基于 VINS,支持双目、GPS 等的 VIO 系统PDFhttps://github.com/HKUST-Aerial-Robotics/VINS-Fusion
OKVISVIO基于 OKVIS 的开源实现PDFhttps://github.com/ethz-asl/okvis
VI-DSOVIODSO 与 IMU 结合的开源 VIO 系统PDF
Page
Video
https://github.com/RonaldSun/VI-Stereo-DSO
ICEBAVIO利用 SLAM 增量特性加速 VIO 的系统PDFhttps://github.com/baidu/ICE-BA

2.2 激光 SLAM

名称标签简介论文地址
LOAMOdometry比较早期的 LOAM 实现,基于 Velodyne-16P 激光
PDFhttps://github.com/laboshinl/loam_velodyne
https://github.com/cuitaixiang/LOAM_NOTED (中文注解版)
A-LOAMOdometry港科大实现的一套基于 ceres 自动求导的 LOAM。相比原版有一些简化但是可读性高适合学习https://github.com/HKUST-Aerial-Robotics/A-LOAM
FLOAMOdometryAn optimized version of A-LOAM and LOAM with the computational cost reduced by up to 3 times.
比 A-LOAM 速度更快的实现
https://github.com/wh200720041/floam
LOAM LivoxOdometry使用 Livox 激光雷达的 LOAM 系统https://github.com/hku-mars/loam_livox
https://github.com/Livox-SDK/livox_mapping
BALMMappingBALM is a basic and simple system to use bundle adjustment (BA) in lidar mapping.PDFhttps://github.com/hku-mars/BALM
LeGO-LOAMOdometryLeGO-LOAM: Lightweight and Ground-Optimized Lidar Odometry and Mapping on Variable TerrainPDFhttps://github.com/RobustFieldAutonomyLab/LeGO-LOAM
https://github.com/facontidavide/LeGO-LOAM-BOR (工程优化版)
M-LOAMOdometryM-LOAM (Multi-LiDAR Odometry and Mapping)
Robust Odometry and Mapping for Multi-LiDAR Systems with Online Extrinsic Calibration
PDFhttps://github.com/gogojjh/M-LOAM
LINSLIOTightly-coupled lidar-inertial odometry and mapping system for ROS compatible UGVs.PDFhttps://github.com/ChaoqinRobotics/LINS---LiDAR-inertial-SLAM
LIO-MappingLIOTightly Coupled 3D Lidar Inertial Odometry and Mapping
比 LIO-SAM 更早的工作
PDFhttps://github.com/hyye/lio-mapping
LIO-SAMLIOLIO-SAM: Tightly-coupled Lidar Inertial Odometry via Smoothing and MappingPDFhttps://github.com/TixiaoShan/LIO-SAM
FAST_LIO2LIOFAST-LIO (Fast LiDAR-Inertial Odometry) is a computationally efficient and robust LiDAR-inertial odometry package. FAST_LIO2
FAST_LIO
https://github.com/hku-mars/FAST_LIO
https://github.com/gisbi-kim/FAST_LIO_SLAM (加闭环)
LLS-LOAMOdometryLidar Odometry and Mapping with Mutiple Metrics Linear Least Square ICPhttps://github.com/YuePanEdward/LLS-LOAM
MULLSOdometryMULLS: Versatile LiDAR SLAM via Multi-metric Linear Least SquarePDFhttps://github.com/YuePanEdward/MULLS
Scan ContextLoop ClosureScan Context: Egocentric Spatial Descriptor for Place Recognition Within 3D Point Cloud MapPDFhttps://github.com/irapkaist/scancontext
Intensity Scan ContextLoop ClosureIntensity Scan Context based Full SLAM Implementation (ISC-LOAM)
在 Scan Context 基础上加入 Intensity 信息
https://github.com/wh200720041/iscloam
OverlapNetLoop ClosureOverlapNet - Loop Closing for 3D LiDAR-based SLAMPDFhttps://github.com/PRBonn/OverlapNet
LCDNetLoop ClosureLCDNet: Deep Loop Closure Detection and Point Cloud Registration for LiDAR SLAMPDFhttps://github.com/robot-learning-freiburg/LCDNet
SSCLoop ClosureSSC: Semantic Scan Context for Large-Scale Place RecognitionPDFhttps://github.com/lilin-hitcrt/SSC

2.3 视觉-激光 SLAM

名称标签简介论文地址
LVI-SAMLVIOLVI-SAM: Tightly-coupled Lidar-Visual-Inertial Odometry
via Smoothing and Mapping
紧耦合的视觉-激光-IMU 里程计系统,基于 Velodyne 激光雷达
PDFhttps://github.com/TixiaoShan/LVI-SAM
R2LIVELVIOA Robust, Real-time, LiDAR-Inertial-Visual tightly-coupled state Estimator and mapping
紧耦合的视觉-激光-IMU 里程计系统,基于 Livox 激光雷达
PDFhttps://github.com/hku-mars/r2live
LIMOLVIOLidar-Monocular Visual Odometry.PDFhttps://github.com/johannes-graeter/limo

2.4 Deep SLAM

3 在线学习

3.1 免费课程

学校课程讲师网址说明
ETHVision Algorithms for Mobile RoboticsProf. Dr. Davide Scaramuzza.http://rpg.ifi.uzh.ch/teaching.htmlScaramuzza 实验室的公开课
台大Mechine Learning李宏毅https://speech.ee.ntu.edu.tw/~hylee/ml/2020-spring.html中文为主,内容相对简单适合入门

3.2 付费培训

网站课程说明
Educative.ioGrokking the Coding Interview: Patterns for coding Questions用模板归类的方式讲解,比刷 LeetCode 要快一些
Educative.ioGrokking the System Design Interview知名的系统设计课程
Brilliant.orgLinear Algebra with Applications解题方式学习,课程很广泛,数学基础课程普遍不错,适合碎片时间学习

Add a Comment

Your email address will not be published.