@ganeshie8
How can I show the shortest distance between two points in R3 space is a straight line?
I know it has to do with \[dS = \sqrt{dx^2+dy^2+dz^2}\] and maybe Euler's equation
I'm thinking we let it depend on the parameter so maybe then we can define our function as \[f(t) = \sqrt{(x'(t))^2+(y'(t))^2+(z'(t))^2}\]
So Euler's equation which is \[\frac{ \partial f }{ \partial y } - \frac{ d }{ dx } \frac{ \partial f }{ \partial y' } = 0\] right, so if we take the partial of f respect to x,y, and z we'll get 0. I think this is working out nicely, I'm going to see what I get as a final result
Are you kindof trying to find a path that minimizes the arc length integral ?
Yup!
This looks interesting
If possible can you put the complete work when u finish
Yeah sure, give me a bit, it might take me a while haha.
I'm trying to do this problem, as I've been learning calculus variation and what not, it's very fascinating stuff.
You gona solve that differential eqn next is it
Find a curve \(r(t)\) such that below arclength integral is minimum : \[ L= \int_a^b \|r'(t)\|dt\]
why not start with showing the 2-D case
yeah im working on 2D case on my notes
kai made this tutorial a while back about euler-lagrange method one of its uses is to minimize integrals like this
I did 2d
\[dS = \sqrt{dx^2+dy^2}\] \[f = \sqrt{1+\left( \frac{ dy }{ dx }\right)^2}\] then you can use Euler's equation, and pretty simple from there, and you will get an equation of a straight line, I'll let you finish it on your own, or if you want the full work just ask. I'm sort of doing the same thing with space
\[\int\limits_{x_1}^{x_2} \sqrt{1+ \left( \frac{ dy }{ dx } \right)^2dx}\] we let S be the minimum so we can define the function above, i think that's the hardest part about the problem really
S ^
Neat huh? :P
Oh I just looked at the post dan made, kai actually did it for 2d, pretty much the same exact thing.
Since we're trying to minimize the value of definite integral of \(f(x,y,y')\) between two points, \[df = f_x dx + f_y dy+f_{y'} dy' \implies \dfrac{df}{dx} = f_x + f_y y' + f_{y'} y'' \tag{1}\] eliminating \(y''\) we get \[ \dfrac{d}{dx}\left(f-y'\dfrac{\partial f}{ \partial y' } \right) - \dfrac{\partial f}{\partial x} = 0\]
Ok so for space, \[\frac{ d }{ dt } \frac{ \partial f }{ \partial x' } \implies \frac{ x' }{ \sqrt{x'^2+y'^2+z'^2} } = C\] we can do this with y and z as well, so we get all the constants. \[\frac{ x' }{ C } = \frac{ y' }{ C_2 }\] doing the same thing with z, we can integrate and get a equation for a straight line in R3 I'm sure
nicee
Yeah I think it works out pretty well :P
Quick question, how do I add the dots for derivatives instead?
I think dots are not allowed in math group, you should use primes/subscripts for derivatives/partials :P
\[\huge \dot{y}\]
Yeah well we use dots in physics :P, this is more so a physics question I think?
Oh ok, haha thanks!
but you're in math group hmm
Haha xD
Ok prime is fine
I got more questions, tougher than this...if you want to have a try at them, I'll be trying them later tomorrow
Or I'll just post the questions I get stuck on :P
sure tag me in all questions, i remember doing few of these in my engineering math.. but really don't remember much.. il check out paul site and read some more in the evening
Ok cool :P, I think it's about time I give you tough questions rather than the easy calc ones hehe.
You should try to derive Euler's formula, it will be very useful, especially for these types of problems
i derived it already in my reply above using total derivative do you have a pdf/something thats really short to read ?
http://gyazo.com/0097701d005ec32fdb4761174e36851c this is equivalent to euler eqution
I see, interesting and not really, I've been just reading my book, but I'm sure there are many calculus variation pdfs out there
Join our real-time social learning platform and learn together with your friends!