A ROS package for analyzing IMU performance using Allan variance on stationary data.
imu_utils is a ROS package that analyzes Inertial Measurement Unit (IMU) performance by computing Allan variance from stationary sensor data. It helps quantify noise parameters like white noise and bias instability for gyroscopes and accelerometers, which is essential for evaluating and calibrating IMUs in robotics and navigation systems. The tool processes data collected over extended periods (e.g., two hours) to provide accurate noise characterization.
Robotics engineers, researchers, and developers working with IMUs for navigation, drones, or autonomous systems who need to assess sensor noise and performance.
It offers an open-source, ROS-integrated alternative to proprietary IMU analysis tools, with support for multiple IMU models and frequencies, making it practical for real-world sensor evaluation and calibration.
A ROS package tool to analyze the IMU performance.
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Seamlessly processes data from ROS topics and rosbags, as shown in the launch file examples, making it easy to integrate with standard robotics setups.
Supports analysis of various IMU models like DJI A3, ADIS16448, and xsens-MTI-100 at different frequencies, demonstrated in sample tests with color-coded plots.
Accurately computes Allan variance to extract key noise parameters such as white noise (gyr_n, acc_n) and bias instability (gyr_w, acc_w), based on reference technical reports.
Provides a free, open-source alternative for IMU performance analysis, with code available on GitHub and sample datasets for validation.
Requires Matlab to generate figures from the scripts folder, which is proprietary software and may not be accessible to all users, limiting visualization options.
Necessitates two hours of stationary IMU data, as stated in the README, which can be impractical for quick evaluations or in environments where stationary operation is difficult.
README has spelling errors (e.g., 'refrence') and provides dataset links only to Baidu Netdisk, a Chinese service that may be inaccessible or slow for international users.