2

Here is(are) the question(s).

Let $f(t): (-t_0-\epsilon, t_0 +\epsilon)\to \Bbb R^3$ be a unit speed parametrized curve in $\Bbb R^3$. Suppose that $k(f(t_0)) > 0$, where $k$ is the curvature function. EDIT: I have a feeling the domain should be $(t_0-\epsilon,t_0+\epsilon)$.

Prove there exists a unique unit speed circle $g(t)$ such that $f(t_0) = g(t_0)$ and $ \lim_{t\to t_0} \frac{|f(t)-g(t)|}{(t-t_0)^2}=0$. EDIT: The homework problem says it should be $f(t_0) = g(t_1)$, but it was pointed out that that is probably wrong.

Conclude that for this circle $k(f(t_0)) = k(g(t_0))$.

Find $g(t)$ for $f(t)=(cos(t),sin(t),t)$ and $t_0=\frac \pi 4$.

Really three questions in one. I'm at a total loss. Best I have is that, since curvature at $t_0$ isn't $0$ then $f'(t)$ and $f''(t)$ are not co-incident. EDIT: In fact, they are orthogonal, because $f(t)$ is unit speed. This is much stronger. I think it should be possible to show that the circle must lie in the plane spanned by $f'(t)$ and $f''(t)$.

My gut tells me that the circle must lie in the same plane as the neighbourhood of $t_0$.

I'm really way out of my depth here. Any sort of insight will help me out!

Thanks.

trystero
  • 701
  • Did you mean $f(t_0) = g(t_0)$ ? – bubba Jan 24 '14 at 09:30
  • Your gut is correct. The circle you seek is the so-called "osculating circle". – bubba Jan 24 '14 at 09:31
  • The homework question doesn't say that. I think the homework question may be wrong. I think it has to be $f(t_0) = g(t_0)$ to be able to conclude $k(f(t_0))=k(g(t_0))$. But I posted on here what the question says in case that was the way it was supposed to be. – trystero Jan 24 '14 at 09:33
  • The osculating circle? Thanks! I'll look more into that. I guess it really was supposed to be $f(t_0)=g(t_0)$ – trystero Jan 24 '14 at 09:35
  • And the curve $f$ is a helix. Everyone who teaches beginning differential geometry classes seems to use this as an example. Boring. – bubba Jan 24 '14 at 09:45
  • See http://math.stackexchange.com/questions/647459/construction-of-an-osculating-circle/647738#647738 – Michael Hoppe Jan 24 '14 at 10:23

0 Answers0