Separation at t = t=\infty

Tom and Jerry both have equal top running speeds and are initially at points A A and B , B, respectively, separated by a distance of d d .

They both spot each other and immediately start running at their top speeds. Jerry runs on a straight line perpendicular to the line A B AB and Tom runs in such a way that its velocity always points towards the current location of Jerry.

Let r ( t ) r(t) denote the distance between Tom and Jerry at time t t .

Find lim t r ( t ) \displaystyle \lim_{t \to \infty} r(t) .

0 0 d 2 \frac{d}{2} d d \infty

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

4 solutions

Ed Sirett
Aug 7, 2017

No maths needed for this! Just some reasoning. Jerry can't pull away from Tom as their speeds are the same. So that rules out infinity. Tom will initially get closer to Jerry and Jerry can't ultimately pull away from Tom so that rules out "d". Tom will always have to travel further than Jerry to catch up so that rules out 0. Leaves an answer <d and >0. Which is d/2.

I like this a lot more than the one below it lol

Bob Milph - 3 years, 10 months ago

You haven't actually shown that d/2 is correct, you have simply established that the answer is between 0 and d, which still leaves infinitely many possible answers. I think it's a shame the this question was written as a multiple choice.

Freddie Hand - 2 years, 9 months ago

Log in to reply

Yes, if it hadn't been multiple choice I would have had to do the advanced maths. Nevertheless often real decisions in life are down to choosing from a set of fixed choices.

Ed Sirett - 2 years, 9 months ago
Abhishek Sinha
Jul 27, 2017

Let us denote the coordinate of Tom at time t t by ( x ( t ) , y ( t ) ) \big(x(t), y(t)\big) , where its velocity vector makes an angle θ \theta with the + X +X axis. Then, from the given conditions of the problem, we have d x d t = u cos ( θ ) , d y d t = u sin ( θ ) , y x u t = tan ( θ ) . \frac{dx}{dt}=u \cos(\theta), \frac{dy}{dt}= u \sin(\theta), \frac{y}{x-ut}=\tan(\theta). The last equation gives, x u t = y cot ( θ ) x-ut= y \cot(\theta) Differentiating both sides of the above equation w.r.t. the variable y y , we have d x d y u d t d y = cot ( θ ) y cosec 2 ( θ ) d θ d y . \frac{dx}{dy}-u \frac{dt}{dy}= \cot(\theta) - y \textrm{ cosec}^2(\theta) \frac{d\theta}{dy}. Using the previous set of equations, we have cot ( θ ) cosec ( θ ) = cot ( θ ) y cosec 2 ( θ ) d θ d y . \cot(\theta) - \textrm{ cosec}(\theta) = \cot(\theta) - y \textrm{ cosec}^2(\theta) \frac{d\theta}{dy}. i.e., d y y = cosec ( θ ) d θ . \frac{dy}{y}= \textrm{ cosec}(\theta) d \theta. Integrating and using the boundary condition ( y = d y=-d at θ = π / 2 \theta =\pi/2 ), we have ln ( y d ) = ln ( cosec ( θ ) cot ( θ ) ) = ln ( tan ( θ / 2 ) ) . \ln(\frac{|y|}{d})=\ln(| \textrm{ cosec}(\theta)-\cot(\theta)|)= \ln(\tan(\theta/2)). i.e., y = d ( tan ( θ / 2 ) ) . |y|=d (\tan(\theta/2)). Finally, r ( θ ) = y cosec ( θ ) = d ( tan ( θ / 2 ) ) cosec ( θ ) . r(\theta) = |y|\textrm{ cosec}(\theta)=d (\tan(\theta/2))\textrm{ cosec}(\theta). It is clear that, θ 0 \theta \to 0 as t t \to \infty . Hence, lim t r ( t ) = lim θ 0 r ( θ ) = d 2 . \lim_{t \to \infty} r(t) = \lim_{\theta \to 0} r(\theta) = \frac{d}{2}. \blacksquare

@Abhishek Sinha Sir, isn't this a Radiodrome Curve???

Aaghaz Mahajan - 2 years, 11 months ago
Peter Macgregor
Aug 7, 2017

If Tom is daft he runs from A to B, then turns through a right angle and continues his pursuit in a straight line, then after his turn he will always trail Jerry by exactly d.

If Tom is smart he decides how long he wants to run, and then runs in a straight line towards the spot which Jerry will be at after this time. Pythagoras's theorem then gives the distance between them as

d 2 + ( u t ) 2 u t \sqrt{d^2+(ut)^2}-ut

When u t > > d ut>>d we can use the binomial theorem to approximate this by u t ( 1 + 1 2 ( d u t ) 2 ) u t ut(1+\frac{1}{2}(\frac{d}{ut})^2)-ut which tends to zero as t t \to \infty .

We conclude that since Tom is neither completely daft nor really smart, he follows a curved path that is substantially better than the right-angled dog-leg and poorer than the straight hypotenuse, and so

0 < answer < d 0< \text{answer}<d

The only available choice among the options is d 2 \boxed{\frac{d}{2}}

I liked the way you put it....I thought a same way.

Ashutosh Kapre - 3 years, 8 months ago

I believe you mean d/2, not d=1/2.

Oli Hohman - 3 years, 10 months ago

Log in to reply

thanks - I'll correct it.

Peter Macgregor - 3 years, 10 months ago
Yezi Joy
Aug 8, 2017

There is an accurate solution that does not actually involve calculus except nominally.

Consider J's horizontal x-displacement X relative to T. It is increasing at the rate of u(1-cosA), where A is the angle between T's velocity and horizontal x-axis. The limiting value of X is what we seek.

On the other hand, consider the distance r between J and T. It is being reduced at the rate of u(1-cosA), because while T is approaching J at rate of u, T is escaping from J at rate of cosA.

Because of the same rate, at t=infinity X= the remaining distance = d-X, namely X=d/2.

I had to read it 3 times to understand it, but once I understood it, I saw how beautiful the solution is. Thank you.

Darko Simonovic - 3 years, 9 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...