In the previous article, we turned an unreachable star into a measurable direction.
Now reverse the question.
If you can measure directions toward an unknown point, can those directions tell you where the point is?
Yes—but only when the measurements give enough independent geometric information.
Quick Answer
Start with two known points, A and B. The segment AB is a known baseline.
From A, measure the direction toward an unknown point P. That constrains P to one ray.
From B, measure a second direction toward P. That constrains P to another ray.
Where the two rays intersect, the unknown position is fixed.
known baseline + direction + direction → position
Localization is the geometry of intersecting constraints.
The Structural Question — How Much Information Is Enough to Fix One Position?
This is the deeper idea behind triangulation.
One measurement usually does not tell us one exact point. It tells us a set of possible points.
A direction from A says:
P lies somewhere on this ray.
A second independent direction from B says:
P also lies somewhere on this other ray.
One constraint gives many possibilities.
A second independent constraint can reduce those possibilities to one intersection.
constraint → locus of possibilities → second constraint → intersection → location
This mental model will return later in surveying, navigation, GPS, robotics, and computer vision.
The Baseline Comes First
Triangulation needs an anchored starting geometry.
If points A and B are known, then the distance and orientation of segment AB are known too.
That segment is the baseline.
Without an anchored baseline, we may know the shape of a triangle but not its absolute size or position.
This is why classical geodetic triangulation began by measuring one baseline carefully, then extended positions outward through connected triangles.
Figure 1. A known baseline anchors the geometry. Two measured directions intersect at the unknown point P.
A Small Example — Solve the Triangle Before Solving the Coordinates
Let the baseline be:
AB = 10 units
Suppose the measured angles are:
α = 55° at A
β = 45° at B
The third angle is:
γ = 180° − 55° − 45° = 80°
Now the law of sines gives the distance from A to P:
AP / sin(45°) = 10 / sin(80°)
so:
AP ≈ 7.180
If we place A at (0, 0) and B at (10, 0), the direction from A is 55° above the x-axis.
Therefore:
x = AP cos(55°) ≈ 4.118
y = AP sin(55°) ≈ 5.882
So:
P ≈ (4.118, 5.882)
The original idea is simple: two directions meet.
Trigonometry turns that picture into numbers.
If You Are Standing at P, What Do You Actually Need to Measure?
The title sounds as if you are standing at the unknown point P.
That creates an important distinction.
Classical forward triangulation measures directions from known stations A and B toward P.
If you are instead standing at P and looking toward known landmarks A and B, simply measuring the angle between the two landmarks is not enough to determine one unique position.
Many points can see the same pair of landmarks under the same angle.
To locate yourself from P, you need more information—for example absolute bearings to the known landmarks, another landmark, or another independent measurement.
What matters is not the number of observations alone, but whether they provide independent constraints.
Why Two Rays Are Sometimes a Bad Geometry
In a perfect drawing, any two nonparallel rays meet cleanly.
Real measurements have error.
If the two rays cross at a healthy angle, a small angular error moves the intersection only a modest amount.
If the rays are nearly parallel, a tiny angular error can move the intersection a long distance.
This is a first lesson in geometry quality:
same measurement accuracy + different geometry → different position accuracy
Surveyors therefore care not only about precise instruments, but also about the shape of the observing network.
From One Triangle to a Network
One triangle can locate one point.
Then that new point can become a known point for the next triangle.
Repeat the process and a chain or network of triangles can spread across a region.
The U.S. National Geodetic Survey describes triangulation as a surveying method that uses trigonometry to determine positions on Earth's surface. Large networks linked many survey points into a common spatial framework.
The important change is architectural:
one baseline → one triangle → connected triangles → common map
Snellius and Large-Scale Triangulation
In the early seventeenth century, Willebrord Snellius used a triangulation network in the Netherlands as part of a geodetic measurement of Earth.
His 1617 work connected a measured baseline with a much larger network of observed angles.
This is the same scaling idea we have seen throughout this chapter:
measure a small number of things directly, then let geometry extend the information.
Earlier articles used that habit to estimate height and Earth's circumference.
Here it becomes a system for measuring position.
Python — Solve the Same Triangle
The example can be computed without any external library.
import math
baseline = 10.0
alpha_deg = 55.0
beta_deg = 45.0
gamma_deg = 180.0 - alpha_deg - beta_deg
alpha = math.radians(alpha_deg)
beta = math.radians(beta_deg)
gamma = math.radians(gamma_deg)
AP = baseline * math.sin(beta) / math.sin(gamma)
x = AP * math.cos(alpha)
y = AP * math.sin(alpha)
print("P =", (x, y))The result is approximately:
P = (4.118, 5.882)
Predict first.
Keep the baseline at 10 and β at 45°.
Change α from 55 to 70.
Before running the code, sketch the two rays. Which way should P move?
Triangulation and Trilateration Are Not the Same
These words sound similar because both methods locate an unknown point from known references.
But they use different measurements.
triangulation → mainly angles
trilateration → mainly distances
A direction measurement gives a ray.
A distance measurement from a known point gives a circle in 2D, or a sphere in 3D.
So the deeper common idea is again:
each measurement creates a geometric constraint; position is where the constraints agree
Does GPS Use Triangulation?
Not in the classical angle-based sense used in this article.
GPS satellites broadcast their positions and precise timing information. A receiver uses signal travel time to form distance-like measurements called pseudoranges.
That makes GPS conceptually closer to distance-based trilateration or, more precisely, multilateration.
A classroom diagram may show three distance circles intersecting.
Real GPS positioning is more demanding: a receiver normally needs signals from at least four satellites to solve for three-dimensional position and its own clock offset.
So this article should leave two ideas separate:
classical triangulation: angles locate a point
GPS: signal timing creates distance-like constraints
Why This Follows Naturally from #14
#14 made unreachable objects measurable by turning lines of sight into angles.
#15 takes the next step:
if a direction can be measured, several directions can intersect.
That is the bridge from astronomy to surveying and navigation.
direction → intersecting directions → position
Why This Story Matters
Measuring the world is not only about knowing how large something is.
We also need to know where things are relative to one another.
Triangulation helped turn isolated landmarks into networks of known positions.
Those networks made surveys fit together, supported more reliable maps, and helped create common spatial reference systems.
Position becomes measurable when independent geometric constraints intersect.
Words to Keep
baseline
A known segment that anchors the starting geometry.
triangulation
A positioning method that uses known reference points and measured angles or directions.
ray
A line that starts at one point and continues in one direction.
constraint
A condition that restricts where an unknown point can be.
intersection
A point shared by two or more geometric constraints.
trilateration
A positioning method based primarily on distances to known points.
pseudorange
A GPS distance-like measurement derived from signal travel time and affected by receiver-clock error and other effects.
One Sentence to Keep
A triangle can locate an unknown point because each measured direction restricts where the point can be, and independent constraints meet at a position.
What Should We Ask Next?
Triangulation can create many known positions.
But a map becomes far more useful when every place can be written in one common coordinate language.
That leads to the next question:
Next: How Did People Turn the Earth into a Grid?
Previous: How Did Ancient Astronomers Measure the Unreachable Sky with Angles?
Sources & Further Reading
- NOAA National Geodetic Survey, “Triangulation, Traverse and Horizontal Networks” — historical triangulation as a trigonometric surveying method for determining positions on Earth.
- NOAA National Geodetic Survey, “Legacy Observations” — practical distinction between angle-based triangulation and distance-based trilateration.
- MacTutor History of Mathematics, “Willebrord Snell” — Snellius' seventeenth-century triangulation network and geodetic work.
- NIST, “How Do You Measure Your Location Using GPS?” — GPS signal timing, satellite positions, and why at least four satellites are normally needed for a position fix.
- GPS.gov, “Trilateration” — classroom explanation of distance-based positioning from known satellite locations.
Research note: Reddit, YouTube, and geometry/surveying communities were used as question sensors for recurring confusion—especially “are two points enough?”, “do I need angles or distances?”, “is GPS triangulation?”, and “why can nearly parallel lines give a poor position?”. They were not used as factual authorities.
Mathematical note: the worked example assumes a flat 2D plane with A=(0,0), B=(10,0), and both measured angles pointing to the same side of the baseline. Real surveying also models measurement uncertainty, reference directions, curvature, and network adjustment.