Agents need at least two methods: where to start and what to do each time step. These will be different in each model, so we need a mechanism for telling the agents what to do. One way we could do this is by inheritance, maybe?
Then, the swarm will need to know about these new agents. Can we do this at instantiation time, with swarm then making as many copies of the agent as it needs?
Need to try this and see how it looks. Writing the new agent class needs to be super simple...
Agents need at least two methods: where to start and what to do each time step. These will be different in each model, so we need a mechanism for telling the agents what to do. One way we could do this is by inheritance, maybe?
Then, the swarm will need to know about these new agents. Can we do this at instantiation time, with swarm then making as many copies of the agent as it needs?
Need to try this and see how it looks. Writing the new agent class needs to be super simple...