Orthogonal Matrices and Their Properties
What You'll Learn
You'll discover what makes a matrix orthogonal and why these special matrices preserve the geometry of space.
The Core Idea
An orthogonal matrix is a square matrix Q whose columns form an orthonormal set—that is, the columns are all unit vectors and mutually perpendicular to each other. This special structure gives orthogonal matrices a remarkable defining property:
Q^T Q = I
In other words, the transpose of an orthogonal matrix is also its inverse: Q^T = Q^(-1).
Why This Matters: Geometry Preservation
Orthogonal matrices are geometric guardians. When you multiply a vector by an orthogonal matrix, two fundamental properties remain unchanged:
Length Preservation
For any vector x and orthogonal matrix Q: ||Qx|| = ||x||
The transformed vector has the exact same length as the original. Think of Q as a rigid motion—like rotating or reflecting a ruler without bending it.
Angle Preservation
Orthogonal matrices also preserve angles between vectors. If two vectors meet at 45°, they still meet at 45° after transformation by Q. This is because the inner product is preserved:
(Qx) · (Qy) = x · y
Intuitive Analogy
Imagine a perfectly rigid frame that you can rotate or flip but never stretch, squish, or skew. That's what an orthogonal matrix does to space—it performs rotations and reflections while keeping all distances and angles intact.
Examples in Practice
- Rotation matrices are orthogonal
- Reflection matrices are orthogonal
- The identity matrix is orthogonal
- Products of orthogonal matrices are orthogonal
Key Takeaway: Orthogonal matrices satisfy Q^T Q = I, making them length- and angle-preserving transformations that act like rigid motions in space.