MATHEMATICS ANALYSIS AND APPROACHES

HIGHER LEVEL
EXPLORATION

Exploring Real-world 2D scenarios by modeling their pursuit

curves

Math AA HL Andrea Ferragina

0.Table of contents

0.0 Table of contents

1. Introduction……………………………………………………………………………… 3

1.0 Introduction………………………………………………………………………… 3

1.1 Derivation of fundamental formula………………………………………………… 4

2. Pursuit scenario………………………………………………………………………….. 6

2.0 Pursuit Scenario in 2D………………………………………………………………. 6

2.1 Description ……………………….………………………………………………….. 6

2.2 Calculations……………………………………………………………………….. 7

3. Variables ……………………………………………………………………………… 15

3.0 Variables n and R and their importance …………………………………………… 15

3.1 Finding coordinates of impact, giving values to R and n ………………………….. 16

4. Conclusion……………………………………………………………………………. 20

4.0 Conclusion …………………………………………………………………………. 20

5. Work cited………………………………………………………………………………. 21

2
1. Introduction

1.0 Introduction

Interactions between two bodies were one, the pursuer, tries to reach the other, the

pursuee can be modeled by a curve of pursuit. If we take as an example a lion and a gazelle, or

any other predator and prey relationship, if the prey follows a known path the predator's course

corresponds to the pursuit curve, which is the path followed by the pursuer trying to reach the

pursuee. The pursuee is always on the pursuer's tangent because the trajectories of the pursuer

and the pursuee are parameterized in time, meaning that the predator is always moving in the

direction of the line with the shortest path that connects the two bodies. The main three types of

pursuit curves are lead, pure, and lag (​​Joshi). This exploration will investigate pure pursuit

curves in their real life context. Plenty of applications of pursuit curves in real life exist and most

humans experience them from a very young age such as childhood games like tag or American

Football where one person has to chase another to try and catch them. Another real-life scenario

that has already been mentioned are predator and prey relationships such as a fox and a chicken.

The most fascinating real life application to me of pursuit curves is in the military, they can be

used to calculate the path that a pursuit missile would need to follow in order to intercept a

bombing plane or another missile.

Aerial gunnery techniques have received a lot of attention in recent years, both for

improvement and analysis (Krishnan 1). I chose this topic for my mathematical exploration as it

is relevant to what is happening in the world. Since the beginning of the new Russo-Ukrainian

struggle in the heart of Europe last year, the subject has become more pertinent and has begun to

surface more frequently in my life, being more discussed both in the media and newspapers. In a

3
recent newspaper article by The Guardian I read that the missile that hit Poland on November

22nd of 2022 was not part of a Russian attack, however it was a Ukrainian pursuit defense

missile trying to intercept an imminent Russian bombing on the country. This fact intrigued me

and spurred me out to explore the calculations, logistics and math behind tracking missiles.

1.1 Derivation of fundamental formula

The arc length formula is essential when finding the equation of pursuit of certain

scenarios, in this section we are going to look at computing the arc length of a function as later it

will be used as a fundamental step that will allow us to start the calculations. We want to

determine the length of the continuous function y = f(x) between the points (a, 0) and (b, 0). We

must also assume that in the same interval the derivative is also continuous. First of all to

determine the formula we need to determine the length of the curve. To do this, we'll split the

interval into n equal subintervals of

width Δx each, and we'll use Pi to

represent the position on the curve for

each of the points. Therefore, by

connecting the points with a succession

of straight lines, we can approximate

the curve. Diagram 1 is an illustration

of this scenario for n = 10.

4
 | |
The length of each of these segments is 𝑃𝑖−1𝑃𝑖 so the length of the curve will be approximately

𝑛
| |
𝐿 ≈ ∑ 𝑃𝑖−1𝑃𝑖 . By taking a larger and larger n value we can get the exact value of L, in other
𝑖=1

𝑛
words: |
𝐿 = lim ∑ 𝑃𝑖−1𝑃𝑖 .
𝑛 → ∞ 𝑖=1
|
Let's now better understand how long each of the line segments are. To do so we can use pythagoras as shown:

Where Δ𝑥 = 𝑥𝑖 − 𝑥𝑖−1and Δ𝑦 =𝑦𝑖 − 𝑦𝑖−1 so when substituting:

* *
[ ]
According to the Mean Value Theorem, on the interval 𝑥𝑖 , 𝑥𝑖−1 there is a point 𝑥𝑖 so that, Δ𝑦 = 𝑓'(𝑥𝑖 ) · Δ𝑥

Consequently the length can be rewritten as:

The exact length of the curve L then is:

However this can also be re-written as a definite integral as shown below:

Which is equal to the arc length formula.

5
2. Pursuit Scenario

2.0 Pursuit Scenario in 2D

To understand how pursuit curves can be applied in complex real life situations such as

situations with 3 pursuers chasing each other, we first must consider the simplest case in 2D so

not on a straight line. The problem that is going to be explored is similar to the one considered by

Pierre Bouguer in 1732 (Yoshihara and Ohira 2). In this section of my investigation the equation

for the pursuit curve for a one to one pursuit and evasion problem where the evader moves in a

straight line (parallel to the y-axis) is going to be obtained.

2.1 Description

To describe this problem I opted to use a missile and a plane, where the missile is the

pursuer and the plane the pursuee. Only when the plane is within the square having side length R

can the missile locate and reach it. The initial positions of the different bodies are (Gent 1):

- Plane; Bottom right corner

- City A; Top right corner, straight above the plane. Thus the plane will move only

vertically

- Missile; bottom left corner, next to the plane and on the opposite corner of City

From these positions we can deduce that the distance from the from Missile to City A distance

Plane to City A, as it can be seen in Diagram 2 that also shows the initial positions of the Missile,

Plane and City A.

6
 The objective of this problem is to firstly obtain the equation of the curve that simulates

the shortest path that the missile would need to travel to reach the plane and secondly find if the

missile is going to be able to reach the plane in the prescribed square location of side length R,

given that the speed of both the plane and the missile is known, and so if the missile is going to

be able to reach the plane before it starts bombing

City A. As it was said before both the velocities of

the missile (VM) and the plane (VP) must be

constant, additionally the vector that represents

instantaneous VM must point at the planes

coordinates at all times, using Geogebra I was able

to represent the situation shown in Diagram 2 in the

Cartesian plane also showing the instantaneous VM.

2.2 Calculations

As it is shown in Diagram 3 the plane’s position at t = 0 is (R, 0), since the plane moves

always in along the line x = R at a constant velocity of Vp when t > 0 the plane’s coordinates will

be (R, Vpt). On the other hand the missile’s position at t = 0 is (0, 0), since the missile does not

follow a linear path both its coordinates are going to change over time, consequently at t > 0 the

coordinates of the missile will be (x, y). Knowing these two points for when t > 0 allows the

slope of the tangent at point (x, y) to the pursuit curve to be calculated:

(Equation 1)

7
This calculation is further clarified in the vector diagram below, where the green vector is the

tangent at point (x, y) to the pursuit curve:

If we rearrange equation (1) to make t the subject we obtain:

Additionally, we are aware that the missile has always traveled a distance of VMt along the

pursuit curve. Using the arc length formula, we obtain:

Now we rearrange the equation above to make t the subject:

Since the rearranged equation of the arc length and the slope of the tangent at point (x, y) both

equal t we can now equate them:

Differentiating both sides will remove the integral. Firstly we need use implicit differentiation to

𝑦
differentiate 𝑉𝑀
apply the product rule to differentiate, secondly the product rule can be applied

to differentiate: (in the process to simplify I let ).

8
 𝑑𝑦
We can now replace 𝑑𝑥
with h(x), which allows us to simplify the equation:

Multiply both sides by VP to obtain:

(Equation 2)

𝑉𝑀 𝑉𝑀
Where 𝑛 = 𝑉𝑃
, , because firstly 𝑉𝑃
can not be zero as the two bodies are always

𝑉𝑀
moving, and secondly 𝑉𝑃
can only be positive as the velocity vector of both the plane and the

missile move in the same direction meaning they have the same sign, either both positive or

𝑉𝑀
negative, resulting in 𝑉𝑃
being positive in any case, so .

If we now substitute h(x) with h and rearrange the equation, as so:

We obtain a separable differential equation, we can now integrate both sides of the equation.

9
 - RHS:

1
Since 𝑛
is a constant we can take it out of the integral:

Now we can solve this integral by substitution:

- LHS:

2
To integrate the left hand side we must substitute ℎ with 𝑡𝑎𝑛(𝑢), if ℎ = 𝑡𝑎𝑛(𝑢), 𝑡ℎ𝑒𝑛 𝑑ℎ = 𝑠𝑒𝑐 (𝑢)𝑑𝑢

To solve this integral we can multiply top and bottom by the same factor, effectively not

changing the result:

2
We can now use substitution where 𝑡 = 𝑡𝑎𝑛(𝑢) + 𝑠𝑒𝑐(𝑢) and 𝑑𝑡 = 𝑡𝑎𝑛(𝑢)𝑠𝑒𝑐(𝑢) + 𝑠𝑒𝑐 (𝑢) 𝑑𝑢

and we obtain:

10
Now we can equate the two sides obtaining:

2 2 2
Since 𝑡𝑎𝑛(𝑢) = ℎ and 𝑠𝑒𝑐 (𝑢) = 1 + 𝑡𝑎𝑛 (𝑢) then 𝑠𝑒𝑐(𝑢) = 1 + ℎ . Hence allowing us to

rewrite the equation above as shown below:

Although we solved the for the indefinite integral to find the equation of the pursuit curve we must

also find the integration constant, C. To do this we can use the point (0, 0), when t = 0. We must also

𝑑𝑦 𝑑𝑦
recall that ℎ = 𝑑𝑥
and that at t = 0, VM = 𝑑𝑥
= 0.

We can substitute C and rearrange the equation using the properties of logs

If 𝑙𝑛(𝑥) = 0, then 𝑥 = 1, so:

11
Now all we need to do is solve for h, so we divide:

At this point we are going to introduce the new variable p, where , this will simplify

the calculations below:

Solving for h:

Now we can substitute h and p with their original variables:

Integrate both sides with respect to x:

(Equation 3)

12
Now we need to solve for C, since we know the missile starts at (0, 0):

Now we can rearrange and substitute C in equation 3:

(Equation 4)

𝑉𝑀
Where 𝑛 = 𝑉𝑃
. We obtained the solution to equation 2.

13
I used Geogebra to graph the function above in red, that symbolizes the path of the missile, and the

path of the plane in blue showing the impact point, which is the intercept of the two functions and the

location of City A.

To produce this graph I used random values of n and R, in the graph above 𝑛 = 1. 6 and 𝑅 = 10.

This led me to realize the importance of the variables n and R, as the point of impact, and so also

if the missile is going to reach the plane before it reaches City A.

14
 3. Variables

3.0 Variables n and R and their importance

As we have said before R = the length of the side of a square in which the missile is able

to locate the plane, this square as it can be seen in Diagram 2 has located in the top right vertex

𝑉𝑀
City A. On the other hand variable n has many more intricacies. Since 𝑛 = 𝑉𝑃
and both VM and

VP are constant then also n must be constant. We can generalize the outcome of different n values

in three main scenarios as it can be seen by the table below.

General n values Velocities Would the missile Explanation

catch the plane in time

𝑛=1 The velocities would be No Since the distance between the

the same Missile and City A is greater then
the distance between the Plane and
City A, if the two velocities are
the same, the missile will not
shorten the distance between itself
and the plane.

𝑛>1 Missile’s velocity would Depends As it will also be explained later

be greater than the planes in Table 2, if the missile’s velocity
is greater than the one of the
plane, It will still not be certain
that the Missile reaches City A in
time, this happens when the
velocity of the missile is greater,
but by a small amount, compared
to the one of the plane.

0<𝑛 <1 Plane’s velocity would be No If the Plane’s velocity is greater

greater than the missile’s than the missile’s the distance
between the two will increase with
time so the missile would never
intercept the plane.
(Table 1)

15
3.1 Finding coordinates of impact, given the values to R and n

In equation (4) we found a relationship between n, the ratio of the velocities, and R the

length of the side of a square in which the missile is able to locate the plane and also their initial

distance. Since the plane always moves along 𝑥 = 𝑅 the missile is going to impact the plane at a

point P, that has as an x-coordinate always R. This allows me to assume that the point of impact

P is going to have as coordinates (R, y(R)), allowing me to substitute R for x in equation (4) as

shown below:

Simplify:

So point the coordinates of point P are going to be equal to R,( 𝑛𝑅

2
𝑛 −1
)

𝑛𝑅
As it can be seen by the equation of the y-coordinate of point P it is equal to 2 , meaning that
𝑛 −1

there are some values of n for which the missile never comes in contact with the path of the

plane which gives further clarification for what is written in table 1. The values of n for which

𝑛𝑅
2 is undefined are 𝑛 = 1 and 𝑛 =− 1 since we have said before that we can
𝑛 −1

eliminate the latter option. This can also be explained through reasoning since the path traveled

by the missile is longer than the one traveled by the plane if 𝑛 = 1 the missile would never reach

16
the plane, except for when 𝐴 = 0 which means the two start at exactly the same position and at

𝑡 = 0 the missile would have already reached the plane.

3.2 looking at values of n when the initial distance is 20km

To find for which values of 𝑛 > 1 the missile does not hit the plane we must solve the

following inequality:

We can let R be any value such as 20km

Using a GDC we can calculate that: − 1 < 𝑛 <− 0. 61803, 1 < 𝑛 < 1. 61803 of these two

we can ignore the first option because n is a positive integer. So if 1 < 𝑛 < 1. 61803 the

missile won’t hit the plane while if 𝑛 > 1. 61803 it will.

Now we can make a list of scenarios and see if with specific values of n the missile will intercept

the plane before it bombs City A taking into account that once the bomber plane reaches City A

it takes a few seconds to deploy the bombs while the missile explodes immediately as soon as it

reaches the plane.

17
 # n R Will it reach the plane before it bombs At what
City A? y-coordinate

1 1 20 It will never Never reaches

2 1.2 20 Not before the plane starts the bombing 54.55

3 1.5 20 Not before the plane starts the bombing 24

4 1.61803 20 As soon as the bombing is about to start 10

5 1.8 20 Before it even reaches City A 16.07

6 3 20 Before it even reaches City A 7.5

(Table 2)

Sample calculation #3:

All the scenarios got then modeled in Geogebra, graphs 1 to 6 below picture all scenarios that

have been calculated in Table 2:

18
(Diagram 6)

19
4. Conclusion

4.0 Conclusion

The aim of my investigation was to determine the path that a tracking missile needs to

follow to reach a bomber jet before it bombs an imaginary city, City A. The path followed by the

missile was modeled by equating the pursuit curve function (Equation 1) and the arc length

formula. By solving the differential equation I was able to find a function that models the path of

the missile, however variables such as n, the ratio of the velocity of the missile and the plane and

R, the size of the missile’s radar also needed to be taken into account when applying the

theoretical model to a real life situation, and so to figure out if the missile was going to intercept

the plane before it bombarded the city or not. Functions like the one used in this exploration, to

model the path of the missile, are able to accurately predict not only the path of the pursuer but

also the point of contact with the target which thanks to the function was shown to be at

𝑛𝑅
coordinates (𝑅, 2 ) where R = the length of the side of a square in which the missile is able to
𝑛 −1

𝑉𝑀
locate the plane, while n = 𝑉𝑃
. Functions such as the one used in this paper however in real life

situations have some disadvantages and are limited, as their success is very dependant on the

velocity of the pursuer and target, as the latter one can only be intercepted when the speed of the

pursuer is substantially greater than the speed of the pursued.

20
5. Works Cited

Gent, Peter M. “Pursuit Curves.” Henderson State University,

https://www.hsu.edu/uploads/pages/2006-7afpursuit.pdf. Accessed 14 December 2022.

Joshi, Prateek. “The Clairvoyant Curves Of Pursuit – Perpetual Enigma.” Perpetual Enigma, 9

January 2013, https://prateekvjoshi.com/2013/01/09/the-clairvoyant-curves-of-pursuit/.

Accessed 14 December 2022.

Krishnan, S. “Pursuit Courses in Aerial Gunnery.” Defence Science Journal, 26 May 2014,

https://publications.drdo.gov.in/ojs/index.php/dsj/article/view/7539. Accessed 14

December 2022.

Yoshihara, Sota, and Toru Ohira. “Home.” YouTube,

https://iopscience.iop.org/article/10.1088/1742-6596/2207/1/012014/pdf. Accessed 14

December 2022.

21