PSO: Particle Swarm Optimisation

The PSO is one of the most widely used form of Swarm-based Optimisation. From this particular model, there are several derivatives which are
featured in this project.

However, the PSO itself is still worth discussing as it forms the foundation of many aspects of the system.

The original concept of the PSO can be attributed to Kennedy, Eberhart and Shi, who were attempting to model the collective behaviours of a bird flock.

This is a very different approach to say, a Neural Network which attempts to model complex behaviours of a complex organism. In contrast, a PSO is
comprised of many individually simple components whose behaviours are also extremely simple, being able to perform simple moves with limited information.

The Components of a PSO

The components of a PSO can be seen as the following:

• Velocity Update Equation
• Inertia Component: The force of earlier velocities
• Cognitive Component: The attractive force of a particle's best position or memory
• Social Component: The attractive force a particle feels due to better neighbours
• Position Update Process

The above components map particle movement in a multi-dimensional space. Moving through this space is a matter of changing position.
The particle's movement is therefore thought of as a combination of force vectors, like in physics, and these vectors are all considered
when determining the final direction that the particle will move.