[back to home](../Home.md)


## Fundamentale techniques


The most common techniques for locating an object are **triangulation** and **trilateration**. 

**Triangulation** uses three directions to find the 3D position of an object. Technically, two angles are sufficient to know the position of an object but the redundancy of the results is important.

**Trilateration** uses four distances to determine the 3D position of an object. However with three distances, only two points are possible, often on the same vertical axis. It is then often obvious to determine the real position.


<p align="center">
  <img src="https://github.com/SII-Public-Research/localisationTechniques/blob/563a7f99daf2ed16d258c86c587b92c7b2fc4883/images/techniques/tri.png?raw=true"
  alt = "Trilateration_Triangulation"
  width= 100% />
</p>


Other techniques are possible and do not necessarily require several sensors. The preliminary analysis of the environment allows to do localization with often only one sensor. The obvious problem is that a change of environment requires a new calibration of the system. 

Localization **by proximity** is the last classical localization methods. The principle is to have fixed beacons emitting a weak signal. If the signal is received, then we know that the object is approximately close to this beacon. The flaw comes from the method itself: to improve the accuracy, the signal strength must be lowered, and therefore the beacon environment must be denser. 

<p align="center">
<img src="https://github.com/SII-Public-Research/localisationTechniques/blob/4fee810e402ff0b2e941f65db358b224ba038154/images/techniques/techniquesBeacon.png?raw=true"
alt = "Trilateration_Triangulation"/>
</p>


## Data types