The purpose of Mind AI is to build an artificial mind based on some advanced concepts: machine learning, representation and meta representation of concepts, concept reflection, reification (concept to meta concept), and denotation (meta concept to concept), and to explore some new concepts. Interaction with the AI is done via IRC.
The OMCSNet-WordNet project aims to improve the quality of the OMCSNet dataset by using automated processes to map WordNet synonym sets to OMCSNet concepts and import additional semantic linkage data from WordNet. It is based on OMCSNet 1.2, a semantic network and inference toolkit written in Python/Java. OMCSNet currently contains over 280,000 separate pieces of common sense information extracted from the raw OMCS dataset. This project is also based on WordNet, an online lexical reference system that in recent years has become a popular tool for AI researchers.
Player provides a language-independent networked interface to robots and their sensors. Supported devices include Pioneer 2DX robots with sonar, odometry & compass, SICK laser rangefinder, ACTS color vision system, GPS, gripper and wireless communications. Stage provides a population of simulated Player devices. Controllers designed using Stage have been shown to work unchanged on real robots and vice versa. Stage aims for low-fidelity simulation of many devices, rather than perfect models.
METSlib is an object-oriented metaheuristics framework in C++ designed to make implementing or adapting models easy. The model is modular: all the implemented search algorithms can be applied to the same model. METSlib implements the basics of some metaheuristics algorithms, such as Random Restart Local Search, Variable Neighborhood Search, Iterated Local Search, Simulated Annealing, and Tabu Search. For each algorithm, you must implement an objective function, a neighborhood (move manager), and some moves. Tabu Search is one of the fastest ways to generate near-optimal solutions to a wide range of hard combinatorial optimization problems.
Encog is an advanced neural network and bot programming library. It can be used independently either to create neural networks or HTTP bot programs. It also includes classes that combine these two advanced features. It contains classes for Feedforward Neural Networks, Hopfield Neural Networks, and self organizing maps. Training can be accomplished using back-propagation, simulated annealing, and genetic optimization. Additional classes are provided for pruning neural networks. Encog also includes advanced HTTP bot programming features. A multi-threaded spider that can store its workload either in memory on a database is provided. HTML parsing is provided, as well as advanced form and cookie handling.
MARIE is a new design tool for mobile and autonomous robot applications, designed to facilitate the integration of multiple heterogeneous software elements. It is a flexible tool based on a distributed model, thus allowing the realization of an application using one machine or various networked machines, architectures, and platforms.