-1

The line $l_1$ passes through the position vector $i-j-2k$ and is parallel to the vector $3i-4j-2k$.

The line $l_2$ passes through the position vector $(1+5\cos t)i-(1+5\sin t )j-14k$ where $0≤ t≤2\pi$ and is parallel to the vector $15i+8j-3k$.

Show that both lines don't intersect

I have solved it by showing that minimum distance between the two lines is greater than 0 by finding the minimum point of distance function of the perpendicular between them.

However I am interested in a method that involves comparing the vector components of the two lines and then somehow showing that two lines don't intersect. (Maybe by contradiction)

Is there a way to solve it through that method?

mathnoob123
  • 1,373

2 Answers2

1

Suppose we have a line parallel to the vector $\mathbf{A}$ that passes through point $\mathbf{P}$ and a line parallel to vector $\mathbf{B}$ that passes through point $\mathbf{Q}$. For the lines to intersect, it is necessary that they lie in the same plane. This in turn implies the vector $\mathbf{P} - \mathbf{Q}$ also lies in that plane. Three vectors lie in the same plane if and only if the cross product of two of them is orthogonal to the third. Thus, if $(\mathbf{P}-\mathbf{Q})\cdot(\mathbf{A}\times\mathbf{B}) \ne 0$, the lines do not intersect. This condition is probably good enough for your problem.

However, note that $(\mathbf{P}-\mathbf{Q})\cdot(\mathbf{A}\times\mathbf{B}) = 0$ does not imply the lines intersect, as they might be parallel (that is, $\mathbf{A}\times\mathbf{B} = 0$). If this is the case, they will only intersect if $\mathbf{A}$ and $\mathbf{B}$ are also parallel to $\mathbf{P} - \mathbf{Q}$ (in which case the lines coincide).

eyeballfrog
  • 22,485
  • Ohh perfect. I was also confused why do we used the formula Smallest distance between two lines= (Distance between two points)x(normal vector common to two lines)/|normal vector|.You cleared that confusion too. Thank you. – mathnoob123 May 22 '17 at 00:02
0

Zwikker (see below) derives an equation the intersection of two lines in the complex plane. The lines have end points $(z_1,z_2)$ and $(z_3,z_4)$, respectively, and can be thought of as vectors $(z_1-z_2)$ and $(z_3-z_4)$. The point of intersection is given by

$$z=\frac{(z_1z_2^*-z_1^*z_2)(z_3-z_4)-(z_3z_4^*-z_3^*z_4)(z_1-z_2)} {(z_1-z_2)(z_3^*-z_4^*)-(z_1^*-z_2^*)(z_3-z_4)}$$

If the vectors $(z_1-z_2)$ and $(z_3-z_4)$ are parallel, then the denominator equals zero by the criterion of parallellity and then $z=\infty$. Unfortunatley, this does not address the problem you have posed in $\mathbb{R}^3$. However, if the lines are parallel, they must lie on a plane. Perhaps that will help you.

Ref: Zwikker, C. (1968). The Advanced Geometry of Plane Curves and Their Applications, Dover Press.

Cye Waldman
  • 7,524
  • It is a good method but I was hoping for a method in which we would show a contradiction would arise when the components of the vectors are equated. – mathnoob123 May 21 '17 at 23:36
  • @FaiqRaees OK, how about if you normalize both vectors and translate them to the origin and see if they are the same? – Cye Waldman May 21 '17 at 23:48
  • (Apologies, I didn't really understood you) I am aware of (in terms of vector operations) cross products, dot products, basically pre-university math. – mathnoob123 May 21 '17 at 23:55
  • Well, in that case, the cross product ought to do the trick. If the lines parallel the cross is zero, $|a||b|\sin\theta$? – Cye Waldman May 22 '17 at 00:04