Agent Farms is a system for modeling and simulation of complex, multi-agent based systems. The system can be used for creating models of multi-agent systems, interactive and distributed simulation, observation and visualisation of the simulation, and population modification and migration. On those models one can learn about the evolution of strategies in the populations of agents.
|Tags||Scientific/Engineering Artificial Intelligence Desktop Environment GNUstep|
Release Notes: Code cleanup. Many compilation warnings have been fixed. This release should install cleanly, and is ready for testing.
Release Notes: Testing code has been removed from Farmer.app. TestDomain and instructions on how to run the example model have been added. The suite can be built with a single 'make' command.
Release Notes: This is a development snapshot of a complete architectural rewrite. Programming knowledge requirements for model creators have been removed.
Release Notes: This version adds a new FarmsViews framework, removes the dependency on the XY plotting framework, and fixes various bugs.
Release Notes: A farmer module for Views was added. Simulation restarting in Farmer.app was implemented. An integer number Lattice class and Lattice View were added. A few convenience methods for arrays to support agent-based simulations were added. Many simulator-side object memory leaks were fixed. When an exception occurs in the simulator, it will not crash anymore, but the exception is forwarded to the farmer's log. New Farmer's documentation was added.