MI-Calib: Multiple IMUs Spatiotemporal Calibrator

0. Preliminaries

If you use MI-Calib in a scientific publication, please cite the following paper 👇:

• under review

Todo List »

• support more inertial ros messages.

1. Overview

The inertial measurement unit (IMU), as an interoceptive sensor typically providing high-frequency angular velocity and specific force measurements, has been widely exploited for accurate motion estimation in modern robotic applications, such as autonomous navigation and exploration. Recently, there has been a trend of integrating multiple IMUs in inertial systems, which outperforms traditional standalone ones regarding the resiliency of sensor failures and measurement depletion and is commercially affordable due to the significantly reduced price for IMUs. For such multi-IMU systems, accurate and consistent spatiotemporal calibration is inevitably required to perform information fusion in a unified framework. Considering most existing methods generally involve additional aiding sensors in calibration, lacking convenience and usability, we propose MI-Calib: a high-performance multi-IMU spatiotemporal calibrator based on continuous-time batch optimization, which utilizes only raw inertial measurements from multiple IMUs and requires no prior.

Our accompanying videos are now available on YouTube (click below images to open) and Bilibili.

2. Build MI-Calib

2.1 Preparation

• install ROS1 (Ubuntu 20.04 is suggested, Ubuntu 18.04 (ros melodic) is also available):

  sudo apt install ros-noetic-desktop-full
  echo "source /opt/ros/noetic/setup.bash" >> ~/.bashrc
  source ~/.bashrc

  Requirements: ROS1 & C++17 Support

• install Ceres:

  see the GitHub Profile of Ceres library, clone it, compile it, and install it. Make sure that the version of Ceres contains the Manifold module. (Ceres version equal to 2.2.0 or higher than that)

• install Sophus:

  see the GitHub Profile of Sophus library, clone it, compile it, and install it. Set option SOPHUS_USE_BASIC_LOGGING on when compile (cmake) the Sophus library, this would avoid to involve fmt logger dependency (as the following spdlog would use internal fmt too, which may lead to conflict).

• install magic-enum:

  see the GitHub Profile of magic-enum library, clone it, compile it, and install it.

• install Pangolin:

  see the GitHub Profile of Pangolin library, clone it, compile it, and install it.

• install cereal, yaml-cpp, and spdlog:

  sudo apt-get install libcereal-dev
  sudo apt-get install libyaml-cpp-dev
  sudo apt-get install libspdlog-dev

2.2 Clone and Compile MI-Calib

• create a ros workspace if needed and clone MI-Calib to src directory as mi_calib:

  mkdir -p ~/MI-Calib/src
  cd ~/MI-Calib/src
  
  git clone --recursive https://github.com/Unsigned-Long/MI-Calib.git mi_calib

  change directory to 'mi_calib', and run 'build_thirdparty.sh'.

  cd mi_calib
  chmod +x build_thirdparty.sh
  ./build_thirdparty.sh

  this would build 'tiny-viewer' and 'ctraj' libraries.

• Prepare for thirdparty ros packages:

  clone ros packages 'sbg_ros_driver' to 'mi_calib/..' (directory at the same level as mi_calib):

  cd ..
  git clone https://github.com/SBG-Systems/sbg_ros_driver.git

  then change directory to the ros workspace to build this package:

  cd ..
  catkin_make -DCATKIN_WHITELIST_PACKAGES="sbg_driver"

• compile MI-Calib:

  catkin_make -DCATKIN_WHITELIST_PACKAGES=""

3. Launch MI-Calib

3.1 Simulation Test

datasets, launch, result visualization

3.2 Real-world Experiments

datasets, launch, result visualization

3.3 Skip Tutorial

Find a configure file named config-real.yaml in /mi_calib/output or from the open-source datasets below:

# Google Drive
https://drive.google.com/drive/folders/11GTlHyzjSkt6ZXAZs9t6Tmg11Nah2Bx_?usp=sharing

Then change the fields in the configure files to be compatible with your dataset (there are detailed comments for each field). You only need to change a few fields related to io (input and output), perhaps some additional fields related to optimization.

Then give the path of your configuration file to the launch file of MI-Calib named mi-calib-prog.launch in folder mi_calib/launch, Then, we launch 'MI-Calib':

roslaunch mi_calib mi-calib-prog.launch

The corresponding results would be output to the directory you set in the configure file.

Attention:

• Sufficiently excited motion is required for accurate spatiotemporal calibration in MI-Calib!
• Considering that the weak observability of the intrinsic parameters (i.e., non-orthogonal factors, rotational misalign- ments, and biases) in the proposed spatiotemporal calibration, they are required pre-calibrated, which can be accomplished in a separate process. (how to calibrate intrinsics of your IMUs, see Real-world Experiments)