To aid in the development of path integration into the model, a separate NetLogo model file was created. Alongside the diagrams and maths presented by the project supervisor last week, a visit to the Khan Academy website and their presentation of 2D linear transformations went a long way to clarifying the approach to be taken and how to apply it using the NetLogo Matrix extension.
To aid in the investigation of linear transformations and accumulators, a series of NetLogo files were created, each one testing new ideas and calculations until clarification and success were achieved. The final result is a Netlogo model, which will rotate the agent at random angles, but when the switch is clicked to signal the return journey, a direct route is taken back to the start position. This can be seen in the above screenshot, where three agents were moved randomly and a direct route back to the start point was attained when the return-journey? switch was selected.
With this model successfully created, the coming week will involve merging it with the original model, with the inclusion of noise in an effort to represent error accumulation inherent in biological systems. The intention is to also split the snapshots into two different collections; one for foraging snapshots and one for homing snapshots, on the basis that insects tend to perform initial orientation journeys to identify their central location within the environment.