This is the open LiDAR dataset for lifelong SLAM, please refer to the following paper: A General Framework for Lifelong Localization and Mapping in Changing Environment.
The dataset is recorded by ROS recoder, and tested by the following ROS version:
- indigo
- kinetic
- melodic
year-month-date-hour-minute-second_index.bag_filtered.bag Exp: 2020-10-31-6-7-27_176.bag_filtered.bag is recorded on 2020/10/31, 6:7:27. The index of the bag is 176.
Download files from Baidu Pan: https://pan.baidu.com/s/1JTTo76MEJGODd_T-HitPBw (code: ef3h)
These bags include such topics:
- poses under
world
frame: /localization/current_pose or /v5_current_pose - odometry fused by IMU and wheel encoder under
base_odom
frame: /odom - static transform between
base_laser
andbase_link
: /tf_static - dynamic transform between
world
andbase_odom
: /tf - pre denoised 2D LiDAR scan under
base_laser
: /scan - raw denoised 2D LiDAR scan under
base_laser
: /raw_scan - the IMU topic: /gyro
- compressed 3D LiDAR pointclouds: /rslidar_packets and /rslidar_packets_difop
Step 1: Install ROS map_server:
sudo apt-get install ros-<version>-map-server
Step 2: Launch ROS and map server:
roscore
rosrun map_server map_server map.yaml _frame_id:="world"
Step 3: Play the bags:
rosbag play **.bag --clock
Note: For ease of use, we recommend you to use launch file:
roslaunch test_2D.launch
Step1 : Install RS-LIDAR driver:
cp ros_rslidar ~/catkin_ws/src
cd ~/catkin_ws/src
catkin_make
Step 2: Launch driver:
roslaunch test_3D.launch
Step 3: Play ros bag:
rosbag play *.bag --clock
We will upload our new data collected from our robots in real world timelessly! ^_^