Solving the Flocking Problem: Finding A,B,C Meeting Point

  • Thread starter Thread starter chester20080
  • Start date Start date
Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
35 replies · 5K views
chester20080
Messages
56
Reaction score
1
We have three masses (A,B,C), (all the same,m) that each one is on one vertex of an equilateral triangle of a side a.Each mass moves at a constant velocity u all the time.The rule for the motion of the masses is that they every one will always move towards the other in the way A->B->C->A.We have to find WHERE they will meet (our professor said that they will meet somewhere,as this can be proven,but we don't have to show that) and WHEN.The system is initially at rest with the initial positions I mentioned,with any two masses to have a distance a=edge and as soon as we let it free the masses start moving with u=constant (in magnitude) and according to that rule.
I can't find any equations to start from.Obviously if I could determine some direction vectors...But how?I mean what is the physics and the math behind such a phenomenon?Please help!
 
Physics news on Phys.org
If I could have these equastions about t,then I could answer.The question is how to find such equations.How can I translate into math the rule each body is heading to the other?As I imagine the masses will make something like a spiral and then they will meet.
 
But how considering just the symmetry can I solve this?The symmetry is that every mass is far away from every other mass the same distance for every t,except for when they meet.As our professor said and intuitively,they will meet at the center of the triangle,but how can I prove this without any equations?
 
chester20080 said:
But how considering just the symmetry can I solve this?The symmetry is that every mass is far away from every other mass the same distance for every t …

so the shape ABC at any time t is … ? :smile:
 
An equilateral triangle of a side a'<a which decreases every time until they meet (a'=0).So...?
 
They will have the initial magnitude generally (u=constant) but their direction will change every time.
 
The direction of the one will be towards the previous position of the other.
 
Will it be direction1= -(direction2+direction3)?
 
chester20080 said:
Will it be direction1= -(direction2+direction3)?
This is purely a geometry (kinematics) problem, and has nothing to do with physics. Consider using a coordinate system. Here are some possibilities:

(a) Cylindrical coordinates with the origin at one of the masses.
(b) Cartesian coordinates with the origin at one of the masses.
(c) Cylindrical coordinates with the origin at the center of the triangle (i.e., center of the inscribed and circumscribed circles)
(d) Cartesian coordinates with the origin at the center of the triangle.

If you had to choose, which one do you thing would be most convenient to use?

Chet
 
As I am not very familiar with cylindrical coordinates,I would choose d),but I sense the right one is something with cylindrical...
 
Last edited:
Cylindrical coordinates we studied only once in analytic geometry and only the basics,the definition and the basic equations.Nothing else,no examples,no further details.This exercise is from the physics course and our professor told us that it is not a part of the course(no such thing will be in the exams),but gave us this problem nevertheless,just to keep those who are interested in action for the holidays.Who knows?
 
So how do I proceed?
 
chester20080 said:
So how do I proceed?
This problem can be set up in cartesian coordinates also. Let the locations of the three masses at time t be given by the coordinates (x1, y1), (x2, y2), and (x3, y3). What is the equation for a position vector from the origin drawn to each of these three points, in terms of the unit vectors in the x and y directions? What is the equation for the position vector from point 1 to point 2? From point 2 to point 3? From point 3 to point 1? In terms of the unit vectors in the x and y directions, what is the equation for a unit vector pointing from point 1 to point 2? From point 2 to point 3? From point 3 to point 1? Using this result, what is the vector velocity of point 1 relative to the origin? Of point 2 relative to the origin? Of point 3 relative to the origin?

Chet
 
Hello Chet

I would like to do this problem.

Chestermiller said:
This problem can be set up in cartesian coordinates also. Let the locations of the three masses at time t be given by the coordinates (x1, y1), (x2, y2), and (x3, y3).

Chestermiller said:
What is the equation for a position vector from the origin drawn to each of these three points, in terms of the unit vectors in the x and y directions?

$$ \vec{r_1} = x_1\hat{i}+y_1\hat{j} $$

$$ \vec{r_2} = x_2\hat{i}+y_2\hat{j} $$

$$ \vec{r_3} = x_3\hat{i}+y_3\hat{j} $$
Chestermiller said:
What is the equation for the position vector from point 1 to point 2? From point 2 to point 3? From point 3 to point 1?

$$ \vec{r_{21}} = (x_2-x_1)\hat{i}+(y_2-y_1)\hat{j} $$

$$ \vec{r_{32}} = (x_3-x_2)\hat{i}+(y_3-y_2)\hat{j} $$

$$ \vec{r_{13}} = (x_1-x_3)\hat{i}+(y_1-y_3)\hat{j} $$

Chestermiller said:
In terms of the unit vectors in the x and y directions, what is the equation for a unit vector pointing from point 1 to point 2? From point 2 to point 3? From point 3 to point 1?

$$ \hat{r_{21}} =\frac{1}{\sqrt{(x_2-x_1)^2+(y_2-y_1)^2}} (x_2-x_1)\hat{i}+(y_2-y_1)\hat{j} $$

$$ \hat{r_{32}} = \frac{1}{\sqrt{(x_3-x_2)^2+(y_3-y_2)^2}}(x_3-x_2)\hat{i}+(y_3-y_2)\hat{j} $$

$$ \hat{r_{13}} = \frac{1}{\sqrt{(x_1-x_3)^2+(y_1-y_3)^2}}(x_1-x_3)\hat{i}+(y_1-y_3)\hat{j} $$
Chestermiller said:
Using this result, what is the vector velocity of point 1 relative to the origin? Of point 2 relative to the origin? Of point 3 relative to the origin?

$$ \vec{v_1} = \dot{x_1}\hat{i}+\dot{y_1}\hat{j} $$

$$ \vec{v_2} = \dot{x_2}\hat{i}+\dot{y_2}\hat{j} $$

$$ \vec{v_3} = \dot{x_3}\hat{i}+\dot{y_3}\hat{j} $$

Is it correct ?

If yes,what should be the next step ?
 
Hello Tanya!

I suggest using polar coordinates for the problem. Can you write down a few equations for the motion of mass ##m##?

To make things simpler, consider the origin at the centroid of equilateral triangle.
 
I haven't worked much with polar coordinates but I will give a try .

I know a couple of things .

$$ \vec{r} = r\hat{r}$$
$$ \vec{v} = \dot{r}\hat{r}+r\dot{\theta}\hat{\theta}$$

Could you elaborate how you would approach this problem .Where should be the reference line of the angle measured ? How would we take into account that radius(the distance between the particle and the centroid) shrinks and gradually reduce to zero ?
 
Tanya Sharma said:
I haven't worked much with polar coordinates but I will give a try .

I know a couple of things .

$$ \vec{r} = r\hat{r}$$
$$ \vec{v} = \dot{r}\hat{r}+r\dot{\theta}\hat{\theta}$$

Could you elaborate how you would approach this problem .Where should be the reference line of the angle measured ? How would we take into account that radius(the distance between the particle and the centroid) shrinks and gradually reduce to zero ?

Yes, those equations work.

Now, use the fact that the speed of masses is constant. You need one more equation and you can obtain it from the ratio ##\frac{r\dot{\theta}}{\dot{r}}##.

For the reference line, you can assume any initial positions you wish. :)
 
  • Like
Likes   Reactions: 1 person
Pranav-Arora said:
Now, use the fact that the speed of masses is constant.

Do you mean ## u = \sqrt{\dot{r}^2+(r\dot{\theta})^2)} ## ? How should I use it ?
Pranav-Arora said:
You need one more equation and you can obtain it from the ratio ##\frac{r\dot{\theta}}{\dot{r}}##.

What is ##\frac{r\dot{\theta}}{\dot{r}}## ? Is it the ratio of the tangential and radial component of velocity ?
 
Tanya Sharma said:
Do you mean ## u = \sqrt{\dot{r}^2+(r\dot{\theta})^2)} ## ? How should I use it ?
The question states that the speed is constant. You have to use it with the relation you obtain between two components of velocity.
What is ##\frac{r\dot{\theta}}{\dot{r}}## ? Is it the ratio of the tangential and radial component of velocity ?
Yes. :)
 
But what is the relation between the tangential and radial components of velocity ?

I do not understand what we are trying to achieve .Please elaborate the general idea .
 
Tanya Sharma said:
But what is the relation between the tangential and radial components of velocity ?
Do you see that the ratio is ##\tan(5\pi/6)##?
I do not understand what we are trying to achieve .Please elaborate the general idea .
Well...a differential equation. :D
 
Pranav-Arora said:
Do you see that the ratio is ##\tan(5\pi/6)##?

No . How did you get that ?

Pranav-Arora said:
Well...a differential equation. :D

A differential equation in ?
 
Tanya Sharma said:
No . How did you get that ?
I am not sure about how should I explain it. I attached a figure, does it help?

Maybe, look at it this way:
$$\tan\frac{\pi}{6}=\frac{r\dot{\theta}}{-\dot{r}}$$

A differential equation in ?
The one you want. From two equations, you can find a differential equation in either ##r## and ##t## or ##\theta## and ##t##.
 

Attachments

  • masses on equilateral triangle.png
    masses on equilateral triangle.png
    4.7 KB · Views: 507
  • Like
Likes   Reactions: 1 person