I study the book "Geometry topology and physics" by Nakahara. And there is something I misunderstand at page 191.
Here, we compute by using Taylor series, the flow of a vector field $X$ :
$$ \sigma^\mu(t,x) = exp(t \frac{d}{ds}) \sigma^\mu(s,x)|_{s=0}$$
I totally agree with this formula.
But then he says :
The last expression can also be written as : $$ \sigma^\mu(t,x)=exp(tX)x^\mu $$
And then I don't understand...
I tried to see what happens when I derivate succesively $\sigma^\mu(t,x)$ but I get something like this :
$$ \frac{d^2}{dt^2}\sigma^\mu(t,x)|_{t=0}=\frac{d}{dt}X^\mu(\sigma(t,x))|_{t=0}=\frac{\partial X^\mu}{\partial \sigma^\nu}(\sigma(t,x))\frac{\partial \sigma^\nu}{\partial t}(t,x)|_{t=0}=\frac{\partial X^\mu}{\partial \sigma^\nu}(\sigma(t,x))X^\nu (\sigma(t,x))|_{t=0}$$
And I'm stuck.
I would like to understand why the expression of the book is true and also where is my mistake in my derivation.
A recall on definitions :
Given a vector field $X \in \mathbb{\chi}(M)$, we have :
$$ \frac{d}{dt}\sigma^\mu(t,x)=X^\mu(\sigma(t,x))$$ $$ \sigma^\mu(0,x)=x$$
And $\sigma : \mathbb{R} *M \rightarrow M $ is called the flow generated by $X$.