Open Source Intelligent Agents - Page 4

PIQLE is a Platform Implementing Q-LEarning (and other Reinforcement Learning) algorithms in JAVA. Version 2 is a major refactoring. The core data structures and algorithms are in piqle-coreVersion2. Examples are in piqle-examplesVersion2. A complete doc

Moved to code.ros.org

An agent-based dialogue management system that maintains a probabilistic nested belief model and uses it in the choice of dialogue strategies

Game Engine and AI framework for playing No Limit Holdem

The Shapefile2Bin project aim to convert a real map (real city) in Shape format (ArgGIS format) to Bin format (Robocup Rescue format). The map after converted is used for Robocup Rescue Simulator (see more at: Robocup Rescue Simulation Project at SF.net)

Ubiquity is a framework for device agents. It includes system level functionality such as networking, XML message passing, agent discovery, and GUI support for building a system of agents that wrap devices, enabling the creation of smart devices.

The success of the Internet has made it possible to access a large amount of information providers world-wide. However, for the customer it is not easy to cope with this new richness of information, due to the heterogeneity and the missing transpar

Cognitive agent based social simulation toolkit (RBOT+MRS) / production system based on ACT-R (http://act-r.psy.cmu.edu) allows for modelling single actor cognitive experiments (RBOT) and multiple actors in a simulated (semiotic) world (RBOT + MRS).

2OPL (Organization Oriented Programming Language) is a programming language for developing agent organizations. The used abstractions in the language are brute facts, effects of actions, norms and sanction rules. Brute facts are the state of the environment. Effect rules describe how actions change the brute facts. Norms describe which environment states counts as violations. Finally sanction rules can be programmed to respond to norm violations. The current implementation is based on logic programming (Prolog in particular). For those users who are not familiar with Prolog we advise the free ebook Learn Prolog Now (http://www.learnprolognow.org). The interpreter supports three different norm specifications. For more information see the documentation or mail one of the developers.

a sandbox for trying my engine ideas

It's an implementation of the A* algorithm together with a grid processor which pre-processes 8-bit bpm graphic files using a variable resolution and an SDL user interface to test it.

A-Life CAD is a toolkit providing a graphical interface for designing autonomous agents that can be linked against an AI game engine. Though it focuses on creating boid-like behaviours, it can also be used for physical simulation or swarm optimisation.

A chatbot framework currently implementing the ErkiTalk and irc protocol. By specifying parsers for the server's input one can easily create custom bots (if one can program in java of course ;).

This project creates an open-source MTS (Message Transport Service) based based on FIPA (www.fipa.org) document SC00084. The goal is to develop a set of Java classes that allow existing Multi-Agent System developers to use FIPA infrastructure components.

ABSim is an Agent-Object-Relationship (AOR) simulation system based on a Java program library. The development of ABSim has terminated. Its successor is <a href="http://oxygen.informatik.tu-cottbus.de/aor/?q=node/2">AOR-JSim </a>

Machine Learning approach to Go

A platform for setting up autonomic services in a distributed environment. Provides service discovery, service provisioning / usage, autonomic adaptation to the context, mobility, support for supervision and service aggregation, and more.

ACLAnalyser is a powerful tool to validate, verify and debug a MAS (Multi Agent Systems). ACLAnalyser sniff all messages exchanged between agents in a MAS, store them in a data base, and recover them later to get interesting information.

ACT-CV provides an easy to use interface between visual objects on the computer screen (or another video source) and a cognitive modeling tool.

An architecture development environment for distributed agent architectures for virtual and robotic agents with focus on human-robot interaction

Main task is automating the creation of simulation models of complex model. The originalities of proposed software are explicit representation of conceptual model in the concept – attribute – relation form and applying expert system approach to transform designed ontology into agent model. Researchers develop the structure and behavior of agents using declarative approach. Web: https://84.237.20.67/RFFI/Main.php

This project about providing run-time support for developing Decentralized Autonomic Computing systems (Eclipse Innovation Grand). Currently the project contains a automatic guided vehicle (AGV) simulator and an editor and visualisation as eclipse plugin

This is a framework to develop Contexts and Agents and a Container to run them. The Container lets each Agent interact with the Context and other Agents. The agents are deployed in jar files and each one runs in an exclusive sandbox.