Quickstart

This page provides a quick introduction to EnactiveAgents. For instructions on installing EnactiveAgents, head to the Installation page.

Running EnactiveAgents

To run EnactiveAgents, do:

python enactiveagents/enactiveagents.py

This starts the simulation. The simulation can be controlled. The simulation is divided into discrete ticks. Each tick, the following two steps are performed sequentially:

1. Let all agents prepare their next primitive interaction with the current state of the world.
2. One by one, in randomized order, let the agents attempt to enact the prepared primitive interaction.

This subdivision of preparation and enaction is present to ensure all agents have the same information available within a single tick. Note that an interaction of an agent might change the world state, which might have ramifications for the interactions of the subsequent agents within that tick.

To change the simulation, open enactiveagents.py and modify the following line:

experiment_ = experiment.basic.BasicVisionExperiment()

The experiment_ variable should be an object of type experiment.experiment.Experiment. Built-in experiments can be found in experiment/basic.py (experiment.basic module).

An experiment can have various controls assigned to it (experiment.experiment.Experiment.controller()). For example, an experiment might allow you to place an object for the agents to interact with.