zoukankan      html  css  js  c++  java
  • awesome SLAM and robot

    持续更新ing

    关于最新VSLAM进展、语义VSLAM和动态场景下的 VSLAM

    https://github.com/1989Ryan/Semantic_SLAM

    https://github.com/leggedrobotics/darknet_ros

    https://github.com/ros-industrial/industrial_training

    https://github.com/yehengchen/Object-Detection-and-Tracking

    kitti2rosbag

    git clone https://github.com/JotellyBarros/image2rosbag_KITTIodometry
    
    cd image2rosbag_KITTIodometry/
    
    python img2bag_kitti_StereoBag.py /home/mate/Downloads/00/ ~/Downloads/00.bag ~/Downloads/00/times.txt

    动态场景下的VSLAM

    1.对比ORB_SLAM2YOLO-DynaSLAM二者在动态场景下的性能:

    说明:编译ORB_SLAM2YOLO-DynaSLAM系统时,对依赖进行了适当调整,若是碰到类似问题,可参考我的fork版本ORB_SLAM2YOLO-DynaSLAM

    使用TUM_RGB-D Dynamic Objects的fr3_walking_xyz数据集进行了实验,并使用TUM useful tools进行了评估:

    ORB_SLAM2的结果如下:

    python tools/evaluate_ate.py results/CameraTrajectory_fr3_walking_xyz.txt ~/Downloads/datasets/TUM_RGBD/rgbd_dataset_freiburg3_walking_xyz/groundtruth.txt --plot ORB_SLAM2-fr3_walking_xyz.png
    0.730144

    python tools/evaluate_ate.py results/KeyFrameTrajectory_fr3_walking_xyz.txt ~/Downloads/datasets/TUM_RGBD/rgbd_dataset_freiburg3_walking_xyz/groundtruth.txt --plot ORB_SLAM2-Keyframe-fr3_walking_xyz.png
    0.654843

    YOLO-DynaSLAM的结果如下:

    python tools/evaluate_ate.py results/CameraTrajectory_fr3_walking_xyz.txt ~/Downloads/datasets/TUM_RGBD/rgbd_dataset_freiburg3_walking_xyz/groundtruth.txt --plot YOLO-DynaSLAM-fr3_walking_xyz.png
    0.072051

    python tools/evaluate_ate.py results/KeyFrameTrajectory_fr3_walking_xyz.txt ~/Downloads/datasets/TUM_RGBD/rgbd_dataset_freiburg3_walking_xyz/groundtruth.txt --plot YOLO-DynaSLAM-Keyframe-fr3_walking_xyz.png
    0.093899

    结论:

    在动态场景中(以本数据序列为例),YOLO-DynaSLAM的性能优于ORB_SLAM2

     

     

    # for melodic
    sudo apt-get install ros-melodic-geodesy ros-melodic-pcl-ros ros-melodic-nmea-msgs ros-melodic-libg2o
    cd catkin_ws/src
    git clone https://github.com/koide3/ndt_omp.git -b melodic
    git clone https://github.com/SMRT-AIST/fast_gicp.git --recursive
    git clone https://github.com/koide3/hdl_graph_slam
    
    cd .. && catkin_make -DCMAKE_BUILD_TYPE=Release
    

      

     

     

     https://github.com/JimGreenJaguar/hdl_people_tracking/tree/noetic

  • 相关阅读:
    一类分治问题
    [POI2005]DWU-Double-row
    [SCOI2007]降雨量
    [POI2007]ODW-Weights(贪心)
    bzoj3427小P的牧场(斜率优化dp)
    UVA10559 Blocks(区间dp)
    LOJ6089 小Y的背包计数问题(根号优化背包)
    [APIO2008]免费道路(生成树)
    bzoj4383(拓扑排序)
    [HEOI2014]平衡(整数划分数)
  • 原文地址:https://www.cnblogs.com/Jaguar/p/12566677.html
Copyright © 2011-2022 走看看