## Attitude Questions

• ### What is the quaternion?

Quaternions are a 4-parameter representation of attitude just as yaw, pitch, and roll are a 3-parameter representation.  Yaw, pitch, and roll based attitude filters encounter problems commonly referred to as "gimbal lock" at certain orientations (e.g. +/- 90 pitch angle).  We use a quaternion based attitude filter because it is continuous over a full 360 degree range of motion such that there are no limitations on the angles it can compute.  However, the VN-100 has a built-in capability to output yaw, pitch, and roll angles from the VN-100 (the sensor automatically converts from quaternions and outputs yaw, pitch, and roll angles if these are preferred).

• ### What are the reference vectors used in VectorNav products?

The attitude filter on the VN-100 and the VN-200 uses two reference vectors, i.e. the direction and magnitude of (i) gravity down and (ii) magnetic North.

• ### How the attitude is determined in VectorNav products?

The on-board Kalman filter compares the measurements from the three-axis accelerometers to the expected magnitude and direction of gravity down to determine the pitch and roll angles.

The filter also compares the measurements from the magnetometers to the local magnitude and direction of Earth's background magnetic field to determine the heading angle (i.e. yaw angle with respect to Magnetic North).

The filter uses the gyros measurements to smooth-out and provide higher bandwidth attitude estimates.

• ### Are there any attitude estimates limitations?

Gyros of all types (as well as accelerometers) are subject to bias instabilities, in which the zero reading of the gyro will drift over time due to inherent noise properties of the gyros.  The attitude filter on the VN-100 uses the accelerometer and magnetometer measurements to estimate the drift in the gyro zero reading over time, such that the reported gyro readings are compensated for this drift.

• ### What is the Vector Processing Engine?

The Vector Processing Engine (VPE) is the suite of algorithms, digital filtering toolboxes, and other software features that run onboard VectorNav sensors.  VectorNav has developed and refined these software tools as a direct result of supporting the implementation of our inertial sensors into applications ranging anywhere from manned/unmanned aircraft to the monitoring of heavy machinery and even stage props for the entertainment industry.  The components of the VPE include:

• Static (Factory) Calibration
• Dynamic Calibration (Real-time Bias Compensation)