Exercises in ODE

S. Kumaresan
School of Math. and Stat.
University of Hyderabad
Hyderabad 500046
kumaresa@gmail.com

1. The geometric interpretation of a vector field is to attach to each point p ∈ U , the vector
F (p) as a directed line segment emanating from p. A typical physical model for this
concept is the velocity vector F (p) in the domain of a (time-independent) fluid flow.

2. Let f : [a, b] → Rn be continuous. If we write f (t) = (f1 (t), . . . , fn (t)), we define

Z b Z b Z b 
f (t) dt = f1 (t) dt, . . . , fn (t) dt .
a a a

The
R b following observation is very useful and may be taken as the defining equation for
a f (t) dt: Z b  Z b
f (t) dt, v = hf (t), vi dt, for all v ∈ Rn . (1)
a a
Rx
3. We have the fundamental theorem of calculus: g(x) := a f (t) dt is differentiable with
derivative g 0 (x) = f (x) for x ∈ [a, b].
Rb
4. In particular, if f : [a, b] → Rn is a continuously differentiable function, then a f 0 (t) dt =
f (b) − f (a).

5. We have the following fundamental inequality:

Z b Z b
f (t) dt ≤ kf (t)k dt.
a a

6. Let f, fn : [a, b] → RN be a sequence of continuous functions. Write fn = (fn1 , . . . , fnN )

and f = (f1 , . . . , fN ). Then fn → f uniformly on [a, b] iff fnk → fk uniformly on [a, b]
as n → ∞.

7. We have the analogue of Weierstrass M -test. With the notation of the last item, assume
that kfn k ≤ Mn for all n. Then the sequence sn := nj=1 fj converges uniformly to
P

a function s : [a, b] → RN . We say that the series ∞

P
n=1 fn is uniformly convergent on
[a, b] to the function s.
Rb Rb
8. If fn → f uniformly, then a fn → a .

1
 9. Let X and Y be metric spaces. A function f : X → Y is said to be Lipschitz if there
exists L > 0 such that d(f (x1 ), f (x2 )) ≤ Ld(x1 , x2 ) for all x1 , x2 ∈ X. The constant
L is called a Lipschitz constant of f . Note that if L0 > L, then L0 is also a Lipschitz
constant of f and that any Lipschitz function is uniformly continuous.
10. Two most important examples of Lipschitz functions are (i) a linear map A : Rm → Rn
and (ii) a differentiable map f : U ⊂ Rm → Rn with
11. f is said to be locally Lipschitz, if for each x ∈ X, there exists rx > 0 such that the
restriction of f to B(x, rx ) is Lipschitz.
12. The most important examples of locally Lipschitz functions are: C 1 functions with
bounded derivative.
Ex. 1. Show that f (t) = t2 is Lipschitz on any closed and bounded subset of R. Is it
Lipschitz on R?
Ex. 2. Show that f (x) = 1/x is Lipschitz on any interval of the form [δ, ∞) where
δ > 0. Is it Lipschitz on (0, ∞)?
2
Ex. 3. Let f (x, t) = x2 e−t sin t on R := {(x, t) : 0 ≤ x ≤ 2, t ∈ R}. Show that f is
Lipschitz on R.
Ex. 4. Show that any linear map A : Rm → Rn is Lipschitz.
Ex. 5. Show that a differentiable function on a convex open set with bounded deriva-
tives is Lipschitz. (Here the norm on any linear map A is defined as kAk := sup{kAxk :
kxk ≤ 1}.)
Ex. 6. Apply the method of successive approximation to the following initial value
problems:
(a) x0 (t) = x(t) and x(0) = 1.
(b) x0 (t) = x + t and x(0) = 0.
(c) x0 (t) = 4tx(t) and x(0) = 3.
(d) x0 (t) = Ax(t) with x(0) = b where A  n
× n real matrix and b ∈ R .
is an n
1 1
(e) Data as in (d) where n = 2 and A = and x(0) = e1 + e2 .
0 1
 
0 −1
(f) Data as in (d) where A = and x(0) = e1 .
1 0
Ex. 7. Show that f (t) = t2/3 is not Lipschitz on R. Exhibit two distinct solutions of
the IV problem: x0 (t) = 3x2/3 with x(0) = 0.
Ex. 8. Show that the IV problem x0 = x3 and x(0) = 2 has a solution on an interval
of the form (−∞, b) for some b ∈ R. What is the behaviour of x(t) as t → b− ?
Ex. 9. Find the integral curves of the following vector fields on the open sets U ⊂ R2 :
(a) f (x, y) = e1 where e1 = (1, 0) ∈ R2 on U = R2 .
(b) f (x, y) = v, a fixed vector v ∈ R2 on R2 .
(c) f (x, y) = e2 = (0, 1) on U = R2 \ {0}.
(d) f (x, y) = (x, y) on U = R2 .
(e) f (x, y) = (−y, x) on U = R2 .
(f) f (x, y) = (x, −y) on U = R2 .

2
Ex. 10. Extend the basic theorem of ODE to following situation: Let Λ be any set and
let U ⊂ Rn be open. Let f : U × Λ → Rn be uniformly Lipschitz in the first variable:
there exists L > 0 such that kf (x1 , λ) − f (x2 , λ)k ≤ L kx1 − x2 k for all λ ∈ Λ. Thus
the vector field f now depends on a parameter “λ”. For instance, the vector field may
depend on time!

Ex. 11. Solve the IV problem x0 = tx + 2t − t3 with x(0) = 0.