2
$\begingroup$

I am trying to justify a result that I saw that says

Let $f:\mathbb{R}\times \mathbb{R}^n$ be a globally Lipschitz function. Then the Cauchy problem $x'=f(t,x), x(t_0)=x_0$ has a unique solution defined for all $t\in \mathbb{R}$.

Is this going to be true ?

First I saw that if we have a linearly bounded function $f(t,x)$ that is there exists a constant $C$ such that $|f(t,x)|\leq C(1+|x|), \forall (t,x)\in \mathbb{R}\times\mathbb{R}^n $ then the cauchy problem has a unique solution defined for all $t\in \mathbb{R}$.

I saw this by using Gronwall's lemma.Since $x(t)=x_0+\int_{t_0}^tf(s,x(s))ds$ we will have that $|x|\leq |x_0|+\int_{t_0}^t|f(s,x(s))|ds\leq |x_0|+\int_{t_0}^tC(1+|x|)ds\leq |x_0|+C(t-t_0)+C\int_{t_0}^t|x|$. We have that for all $t'\in [0,t]$ there exists $T$ such that $T\geq t'$ and so $C(t-t_0)\leq C(T-T_0)$. Letting $|x_0|+C(T-t_0)=M$, we have $|x(t)|\leq M+\int_{t_0}^t|x|ds$, and now using Gronwall's lemma we have that $|x(t)|\leq Me^{C(t-t_0)}$, and so the function is defined $\forall t\in \mathbb{R}$. Now suppose the maximal interval of existence $(\alpha,\beta)$ isn't $\mathbb{R}$, without loss of generality we can assume that $\beta <\infty$. We would have $\lim_{x\rightarrow \beta}x(t) $ will be in the border of $\mathbb{R}^n$, this is a contradiction since we know that $x(t)\in \mathbb{R}^n, \forall t\in \mathbb{R}$.

Now with this I have tried to show that a globally Lipschitz function is Linearly bounded but all I got is that if we assume that $f(t,x)=f(t+1,x), \forall (t,x)\in \mathbb{R}\times \mathbb{R}^n$ , then I can prove that it is Linearly bounded. I can't seem to do this for the more general case, any help is aprecciated. Thanks in advance.

New edit : I think I was able to see in the general case but want to make sure this works, I am not interely sure it does because my constant is changing. Any $t$ is a compact interval of the form $[0,T']$. We have that $|f(t,x)|-|f(t,x_0)|\leq |f(t,x)-f(t,x_0)|\leq L|x-x_0|\leq L|x|+L|x_0|$. So we have that $|f(t,x)|\leq |f(t,x_0)| +L|x|+L|x_0| \leq C(1+|x|)$, but here we have that $C$ is changing if choose another $t$ the maximum of the function $|f(t,x_0)|$ can change.

$\endgroup$
1
$\begingroup$

As you did in the periodic case or similarly over the time interval $[0,1]$, you can construct a linear bound over any time interval $[-N,N]$. The coefficients may grow with $N$, but by the continuity you know that the bounds exist. Uniqueness tells us that the restriction of the $N+1$ solution to the $N$ interval has to coincide with the $N$ solution.

See also https://math.stackexchange.com/questions/347599/questions-about-the-picard-lindelöf-theorem and the links there for all-at-once proof ideas.

$\endgroup$
1
  • 1
    $\begingroup$ Ah ok now I got it , the coefficients may change and so by that reason we consider the funciton only on each interval of the compact covering of $\mathbb{R}$. Thanks. $\endgroup$
    – Something
    Oct 14 '20 at 13:14

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.