You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello, I'm trying to estimate the orientation using Madgwick filter in a simulated rosbag created with Gazebo, but the estimation has some errors compared to the ground-truth. I compare the results using https://github.com/MichaelGrupp/evo. Here is the graphic of the error in degrees:
I tried with different configurations and parameters values but this is the best I could get. As we can see, it has a maximum error of 8.8° and a mean of 4.2°. Is it normal to get these errors? or should it be something more precise?
Hello, I'm trying to estimate the orientation using Madgwick filter in a simulated rosbag created with Gazebo, but the estimation has some errors compared to the ground-truth. I compare the results using https://github.com/MichaelGrupp/evo. Here is the graphic of the error in degrees:
I tried with different configurations and parameters values but this is the best I could get. As we can see, it has a maximum error of 8.8° and a mean of 4.2°. Is it normal to get these errors? or should it be something more precise?
The IMU is simulated with the Intel RealSense Gazebo ROS plugin (https://github.com/pal-robotics/realsense_gazebo_plugin) and the magnetometer with Hector Gazebo Ros Magnetic plugin (https://github.com/tu-darmstadt-ros-pkg/hector_gazebo)
The sensor configuration:
The text was updated successfully, but these errors were encountered: