Allan ROS2

ROS2 package for analysing noise parameters in IMU's using allan deviation. Designed for use with Kalibr and PX4.

Note : This package is under development. If you find any bugs or errors create an issue, or even better a PR fixing it.

Features

1. Outputs noise parameters file for use with Kalibr, Follows the Kalibr IMU Noise Model.
2. Supports sensor_msgs/Imu, px4_msgs/SensorCombined, px4_msgs/VehicleImuStatus to read PX4-ROS2 bags, No conversion required.

Usage

This package has been tested with ROS2 Foxy and Humble

Build

   # create a new ros2 workspace 
   mkdir ~/allan_ws && cd ~/allan_ws
   mkdir src && cd src
   
   # clone 'px4_msgs' 
   git clone https://github.com/PX4/px4_msgs.git
   
   # clone the package
   git clone https://github.com/CruxDevStuff/allan_ros2
   
   cd ~/allan_ws
   
   # install dependencies
   rosdep install --from-paths src -y --ignore-src
   
   # build the workspace
   colcon build 

Configure

Populate the parameters in ~/allan_ws/src/allan_ros2/config/config.yaml with the appropriate values for you bag. example configuration files

allan_node:
  ros__parameters:
     topic: # /your/topic 
     bag_path: # path/to/bag.db3
     msg_type: # use 'ros' / 'px4_sensorcombined' / 'px4_vehicleimustatus' 
     publish_rate: # imu publish rate (hz) 
     sample_rate: # rate to sample data from bag. Higher sample rates take longer to compute 

Note : Requires rebuilding after configuration colcon build --packages-select allan_ros2. This will be fixed in a future release.

Run

Launch the node

   # requires building after changing config
   colcon build --packages-select allan_ros2
   
   # source the workspace 
   source ~/allan_ws/install/setup.bash
   
   # launch the node 
   ros2 launch allan_ros2 allan_node.py

Output

The node outputs deviation.csv and contains the computed raw deviation values for all 6 axis.

Launch output

[allan_node-1] [INFO] [1673168790.110605813] [allan_node]: IMU Topic set to : # topic in config.yaml
[allan_node-1] [INFO] [1673168790.110669795] [allan_node]: Sample rate set to : # sample rate in config.yaml
[allan_node-1] [INFO] [1673168790.111149862] [rosbag2_storage]: Opened database # path set in config.yaml
[allan_node-1] [INFO] [1673168793.933100186] [allan_node]: Bag length (Seconds): # detected bag length 
[allan_node-1] [INFO] [1673168793.933151985] [allan_node]: Sampling data from bag...
[allan_node-1] [INFO] [1673168799.354297658] [allan_node]: Total samples : # samples taken from bag 
[allan_node-1] [INFO] [1673168799.354340691] [allan_node]: Computing variance and deviation
[allan_node-1] [INFO] [1673168804.424160542] [allan_node]: DONE, deviation output logged to 'deviation.csv'

Example deviation.csv

   # sample period     # accel x    # accel y    # accel z    # gyro x     # gyro y     # gyro z
78.10000000000000853, 0.0007594614, 0.003055848, 0.001219371, 0.001133672, 0.001355157, 0.001293785
78.20000000000000284, 0.0007684184, 0.003033959, 0.001216356, 0.001142654, 0.001362136, 0.001301119
78.30000000000001137, 0.0007739545, 0.002989617, 0.001177716, 0.001129241, 0.001408807, 0.001358467
78.40000000000000568, 0.0007518549, 0.003035895, 0.001149335, 0.00116584,  0.00140981,  0.001331341

Noise parameters

To obtain the noise paramaters run analysis.py with deviation data. analysis.py outputs imu.yaml and deviation plots. imu.yaml contains the random_walk and noise_density values for the IMU and follows the kalibr IMU Noise Model

python3 src/allan_ros2/scripts/analysis.py --data deviation.csv

Example imu.yaml

#Accelerometers
accelerometer_noise_density: 1.86e-03   # noise density (continuous-time)
accelerometer_random_walk:   4.33e-04   # bias random walk

#Gyroscopes
gyroscope_noise_density:     1.87e-04   # noise density (continuous-time)
gyroscope_random_walk:       2.66e-05   # bias random walk

rostopic:                               # fill your imu topic
update_rate:                            # fill your imu publish rate (hz)

Plots

Acknowledgements

Thanks to Russell Buchanan. Computation code adapted from allan_variance_ros