Our project supports up to 4 separate optimisers at once each running in a separate window. This allows a user to perform collective experiments by testing up to 4 separate approaches to a given objective function.

The program supports adjustable graphical rendering of the visualisations including:

• Resolution
• Interpolation

The project offers a wide variety of swarm-based optimisation strategies and 24 objective functions. Furthermore, the project is highly modular. This modularity is realised in the project structure that enables anyone to extend it in the following ways:

• Adding new Objective Functions
• Adding new Optimiser Strategies

This capacity to add new functions and strategies means that the target audience, researchers and academics, can develop their own functions and strategies and then implement it into the system and most importantly, visualise the consequences and performance of this. This extensibility really makes the project a great tool for use in both academia and teaching of artificial intelligence.

The viewing environment for the optimisation visualisation is not merely static. Rather it resides in a skybox and the user has a first person perspective of the landscape, able to zoom in/out, rotate their view and shift their perspective in real time.

The user also has the ability to perform mathematical transformations to the domains of the objective functions, giving them freedom to experiment with different aspects of the landscapes withot requiring new domains to be specified internally in the system. Transformations are a powerful feature that vastly extend the usability and flexibility of the system without requiring proportional code changes or additions.

A point of extension to add additional capacity to the project is in the area of VR. One of the ways that the project could be extended is by configuring the project to integrate with a VR system so that the visualisation is presented over a VR headset as opposed to a traditional screen. This kind of capacity would have several positive effects to the user experience.

• Increased immersion
• Greater viewing capacity
• Increased understanding of spacial nature of the fitness landscape

Another area in which the project could be extended could be to include for higher dimensional functions and landscapes, generalising the 1 or 2 dimensional system to an N dimensional one. This would include the provision of higher dimensional mapping strategies to visual n-dimensional problems more easily.