Let this image be in the x,y(x) plane. I am trying to compute the minimum time taken to get from A to B. I have determined that, using Euler-Lagrange methods,$$t=\sqrt{\frac{R}{g}}\int_{x_A}^{x_B}\frac{\sqrt{1+y'^2}}{\sqrt{R^2-x^2-y^2}}dx,$$ where $R$ is the radius of the circle and $g$ is acceleration due to gravity. The Euler-Lagrange is given by $$\frac{\partial F}{\partial x}-\frac{d}{dy}\Big(\frac{\partial F}{\partial x'}\Big)=0.$$ I have computed that $$\frac{\partial F}{\partial x}=\frac{x\sqrt{1+y'^2}}{(R^2-y^2-x^2)^{3/2}},$$ but I am struggling to see how to compute the rest. Is $\frac{\partial F}{\partial x'}=0$? This would make sense, except I know that the solution is as follows: $$x(\theta)=R\Big[(1-b)\cos{\theta}+b\cos{\frac{1-b}{b}\theta}\Big],$$ $$y(\theta)=R\Big[(1-b)\sin{\theta}-b\sin{\frac{1-b}{b}\theta} \Big],$$ with $b \in [0,1]$. I am struggling to see how this would be obtained. The book says that the calculation is tedious, but can someone please show me how to do this?
Asked
Active
Viewed 196 times
3
-
1You mixed up the equation. It's F_x - d/dx(f_y')=0 – Zaros Mar 22 '16 at 16:21
-
Still struggling to solve for the final answer. Any suggestions? – ODP Mar 22 '16 at 17:53
-
I would try to convert it to polar. I bet that's way easier. I'll try to work through it but the differential equation is horrid. Second order non-linear differential equations are obnoxious to deal with. – Zaros Mar 22 '16 at 18:00
-
OK thanks, I will keep trying – ODP Mar 22 '16 at 18:04
