IMU tools for ROS

Overview

IMU-related filters and visualizers. The repository contains:

• imu_filter_madgwick: a filter which fuses angular velocities, accelerations, and (optionally) magnetic readings from a generic IMU device into an orientation. Based on the work of 1.

• imu_complementary_filter: a filter which fuses angular velocities, accelerations, and (optionally) magnetic readings from a generic IMU device into an orientation quaternion using a novel approach based on a complementary fusion. Based on the work of 2.

• rviz_imu_plugin a plugin for rviz which displays sensor_msgs::Imu messages

Installing

From binaries

This repo has been released into all current ROS1 and ROS2 distros. To install, simply:

sudo apt-get install ros-<YOUR_ROSDISTO>-imu-tools

From source (ROS1)

Create a catkin workspace (e.g., ~/catkin_ws/) and source the devel/setup.bash file.

Make sure you have git installed:

sudo apt-get install git

Clone this repository into your catkin workspace (e.g., ~/catin_ws/src; use the proper branch for your distro, e.g., melodic, noetic, ...):

git clone -b <YOUR_ROSDISTO> https://github.com/CCNYRoboticsLab/imu_tools.git

Install any dependencies using rosdep.

rosdep install imu_tools

Compile the stack:

cd ~/catkin_ws
catkin_make

From source (ROS2)

Follow the steps from the ROS2 Creating a workspace documentation, but instead of cloning the sample repo, clone the proper branch of this repo instead:

git clone -b <YOUR_ROSDISTO> https://github.com/CCNYRoboticsLab/imu_tools.git

More info

All nodes, topics and parameters are documented on this repo's ROS wiki page.

What's next after filtering your IMU?

Once you have a filtered orientation from imu_filter_madgwick or imu_complementary_filter, a common next step is fusing it with wheel odometry (and optionally GPS) to produce a full odom -> base_link state estimate. Two ROS 2 packages that do this:

• robot_localization: well-established EKF/UKF state estimator, widely used across the ROS ecosystem.

• FusionCore: UKF with online gyro bias estimation (yaw drift compounds slower over long runs), native GPS fusion for outdoor robots, and a numerically stable filter that handles aggressive maneuvers without diverging. Available on apt for Jazzy and Humble (ros-jazzy-fusioncore).

Both take a filtered IMU topic as input.

pre-commit formatting checks

This repo has a pre-commit check that runs in CI. You can use this locally and set it up to run automatically before you commit something. To install, use apt:

sudo apt install pre-commit

To run over all the files in the repo manually:

pre-commit run -a

To run pre-commit automatically before committing in the local repo, install the git hooks:

pre-commit install

License

• imu_filter_madgwick: currently licensed as GPL, following the original implementation

• imu_complementary_filter: BSD

• rviz_imu_plugin: BSD