Foundations — Transformation between frames — direction cosine matrices

This page builds — from absolutely nothing — every symbol, word, and picture the parent note Transformation between frames — DCMs leans on. If you have ever felt a formula there "appear from nowhere", the missing piece is on this page.

1. A vector — the arrow that stays put

Picture a physical arrow floating in the room: the rocket's thrust. It has a size (how many newtons) and a way it points. That arrow is real and fixed — it does not care how you set up your rulers.

2. Axes and a frame — the rulers you measure against

The little hat $\hat{}$ means "unit vector": an arrow of length exactly $1$ that only carries direction, not size. Think of it as the tip of a ruler pointing one way. The subscript $1,2,3$ just numbers them (many books write $x,y,z$ — same thing).

3. Components — the three numbers of an arrow

The stacked column of three numbers is written ${\mathbf{v}}^A=\left(\begin{array}{c}{v}_1^A\\ v_2^A\\ v_3^A\end{array}\right)$. Bold-upright $\mathbf{v}$ means "the list of numbers", distinct from the arrow $\vec{v}$ itself. The parent note's whole equation ${\mathbf{v}}^B={\mathbf{C}}^{BA}{\mathbf{v}}^A$ is a rule for turning one column of numbers into another column — same arrow, new frame.

4. The dot product — the tool that measures "how much along"

We need one specific question answered: given the arrow, how many steps does it reach along a chosen axis? The tool that answers exactly this is the dot product — and this is why the topic reaches for it and not, say, a cross product (which measures area/perpendicularity, the wrong question here).

Two magic facts follow, and the whole DCM rests on them:

The first line is orthonormality: same axis gives shadow $1$, perpendicular axes give shadow $0$. Because of it, dotting the arrow with ${\hat{a}}_i$ cleanly plucks out the $i$-th component and ignores all the others. This is the single trick the parent note uses to derive every matrix entry.

5. Cosine — why a cosine, and what it reads off

Why does the topic live and breathe cosine? Because the dot product of two unit arrows is a pure cosine: ${\hat{b}}_i\cdot {\hat{a}}_j=1\cdot 1\cdot \cos {\theta }_{ij}=\cos {\theta }_{ij}.$ So each "how twisted is axis ${\hat{b}}_i$ from axis ${\hat{a}}_j$" question is answered by one cosine. That is literally why the matrix is called the direction cosine matrix.

6. Angle between axes, and the twist that relates two frames

When frame $B$ is turned relative to frame $A$, every $B$-axis makes some angle with every $A$-axis. There are $3\times 3=9$ such angles, ${\theta }_{ij}$ = angle between ${\hat{b}}_i$ and ${\hat{a}}_j$.

7. A matrix — a grid of numbers that acts on a column

The symbol ${\sum }_j$ ("sum over $j$") just means "add up as $j$ runs $1,2,3$" — a compact way to write the three-term sum on the right. In pictures: row $i$ of the matrix dots into the column, and that shadow becomes the $i$-th output number. So a matrix is nothing but three dot products stacked — which is exactly why the DCM (built from dot products) is a matrix.

8. Transpose, identity, and inverse — flipping and undoing

9. Right-handed frames and the determinant sign

The determinant $\det \mathbf{C}$ is a single number measuring how a grid scales volume and whether it flips handedness. For a pure rotation between two right-handed frames it equals $+1$ (volume preserved, handedness preserved). A value of $-1$ would mean a mirror flip — never a real rotation. This is exactly the parent note's "$\det =+1$, proper rotation" claim.

Prerequisite map

Read it upward: arrows and rulers give components; the dot product plus cosine give one direction-cosine entry; nine of them stacked as a matrix form the DCM; transpose and handedness give its inverse and its $+1$ determinant.

Equipment checklist

Reveal each only after you can say it out loud:

When every line above answers instantly, you are ready for the parent note Transformation between frames — DCMs.