Overview

Warning: Currently in Alpha

This package launches multiple sensors on a backpack system. With launch, the nodes of the sensors will start and it will open a rviz window to show each sensor and coordinate systems (CS). Additionally, the sensor data can be saved in bag files, default setting is set to false.

In more detail: The launch file backpack.launch includes the backpack_sensors.launch and also starts the static CS transformations between the sensors, and optionally the SLAM calculations based on the Lidar data. The latter receives GPS information. Lastly a RVIZ window can be launched.

The launch file backpack_sensors.launch starts the nodes for the Velodyne lidar and the Xsens/Microstrain IMU each in their namespaces for better understanding of the affiliations when listing active topics.

The currend Includes lidar sensor Velodyne VLP16 and IMU sensor Microstrain 3DM-GQ7 and the past version IMU Xsens GTI-700. It was tested on Ubuntu 20.04 and ROS noetic.

The backpack also includes a multispectral camera from MicaSense. Communication scripts can be found under "src".

Overview of the hardware connections of the backpack system this code was based on:

Overview of the connections between the ROS packages and nodes. Note that "lidar1" is used as example and was also implemented the same way for lidar2.

Visualisation of the backpack that the documents were written and customised for.

Dependencies

please install everything with

cd your/catkin_ws
rosdep check backpack_pkg #to see what is needed
rosdep install backpack_pkg

Along some standard packages, you will need the following. These will also be installed with rosdep.

LiDAR

• velodyne for ROS noetic more information here http://wiki.ros.org/velodyne http://wiki.ros.org/velodyne/Tutorials/Getting%20Started%20with%20the%20Velodyne%20VLP16

IMU

• microstrain_inertial_driver for ROS noetic more information https://github.com/LORD-MicroStrain/microstrain_inertial

OR

• xsens_ros_mti_driver for ROS noetic more information http://wiki.ros.org/xsens_mti_driver

SLAM

• StaticMapping https://github.com/EdwardLiuyc/StaticMapping

OR

• FastLIO2 https://github.com/hku-mars/FAST_LIO

Instructions

to launch package including navigationa and connected sensors, while also recording data in bag file, run the following in directory of catkin workspace:

$ source devel/setup.bash
$ roslaunch backpack_pkg backpack.launch record:=true

to launch sensors and not record data in bag file:

$ source devel/setup.bash
$ roslaunch backpack_pkg backpack_sensors.launch

bag files will be saved in backpack_pkg/bagfiles with their date and time, such as: backp_imu_lidar_2021-12-22-13-29-33.bag

to play a bag file launch without recording and type in a new terminal:

$ rosbag play <path/to/rosbag-name>.bag

example:

$ rosbag play rosbag_src/bagfiles/backp_imu_lidar_2021-12-22-13-29-33.bag

more information: http://wiki.ros.org/rosbag/Commandline

After creating a catkin workspace, installing the other to packages and downloading this repo, make sure to

• activate the backpack_pkg by using catkin-make pkg backpack_pkg
• go back to the home of your catkin workspace and use "catkin_make" and "rosdep update"

Paper

• Segmentation of Forest Vegetation Layers Based on Geometric Features Extracted from 3D Point Clouds: Goebel et al. (2023)

• Backpack System for Capturing 3D Point Clouds of Forests, postprint: Goebel & Iwaszczuk (2023)

Contact

If you do not have a GitLab account, you can write me any issues, ideas and comments to: mona.goebel@tu-darmstadt.de