If $g ∈ \mathrm{SO}(3)$ is the rotation about axis $p$ by angle $α$, and $h$ is a rotation mapping $p$ to another line $q$, then $g$ conjugated by $h$ is the rotation about $q$ by the same angle $α$. I've seen this fact presented as obvious, but I don't find 3D rotations trivial to visualize in general, and would like to see a formal proof. This could presumably be brute-forced in a number of ways, but I'm interested in an argument that avoids lengthy computation.

Since $hgh^{-1}$ obviously fixes $q$, that is indeed the axis of the resulting rotation, so it would suffice to prove it rotates $q^{\bot}$ by $α$.


An easy way to see this is to use Rodrigues's rotation formula: $$ g{\bf v}=\cos \alpha \,{\bf v}+(1-\cos \alpha)({\bf u}\cdot {\bf v}){\bf u}+\sin \alpha \,{\bf u}\times {\bf v} $$

where ${\bf u}$ is the axis of rotation. Now it can be shown that conjugating $g$ by $h$ gives $$ hgh^{-1}{\bf v}=\cos \alpha \,{\bf v}+(1-\cos \alpha)((h{\bf u})\cdot {\bf v})h{\bf u}+\sin \alpha \,(h{\bf u})\times {\bf v}. $$

This is a rotation by $\alpha$ about axis $h{\bf u}$.

| cite | improve this answer | |
  • $\begingroup$ Thanks for the answer. I was aware of the formula, but was hoping there was a more conceptual proof. I'll wait a bit more before accepting this just in case, but I doubt anyone else will comment. $\endgroup$ – user54748 Feb 8 '16 at 21:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.