Integrated Navigation Test 2 Practice

A quaternion represents attitude by encoding a 3D rotation into a four-parameter unit quaternion q = [q0, q1, q2, q3], where q0 is the scalar part and (q1, q2, q3) is the vector part, with the unit-norm constraint q0^2 + q1^2 + q2^2 + q3^2 = 1. This single object captures orientation without relying on a sequence of axis rotations, which is what Euler angles do. The main advantage is the absence of gimbal lock. Euler angles describe orientation as consecutive rotations about axes, and when the pitch approaches ±90 degrees, two axes align and one degree of freedom is lost. Quaternions avoid this singularity because they encode rotation in a way that does not depend on a specific sequence of axis rotations. Rotations are composed simply by quaternion multiplication. If you apply one rotation and then another, you multiply their quaternions to get the combined rotation. This makes the math stable and straightforward for iterative updates, which is especially important when you’re chaining many small rotations in real time. For attitude estimation using filters like an EKF, quaternions offer numerical robustness. They avoid the singularities that come with Euler angles, allow smooth interpolation between orientations (via operations like SLERP), and maintain a consistent parameterization of rotation. Normalization is easy to enforce to keep the unit-norm constraint, helping prevent drift from accumulating. In contrast, the idea that a quaternion is a scalar is incorrect, since a quaternion has four components. It doesn’t use Euler angles “under the hood,” though you can convert between representations if needed; it is a distinct, parallel representation for 3D rotations. And yes, quaternions can represent 3D rotations, so that last option doesn’t fit.