• Projects

    Note that this page is under construction.

    In the past, we have been developing and researching artificial life of these organisms, ranging from walking mobile robots to Internet agents enabling e-commerce. Multidisciplinary fields such as ethology, psychology, sociology and mathematics provide us necessary knowledge to incorporate intelligence and behavior of existing biological systems into these organisms. Many of research projects finished under the leadership of Doc. Ing. Pavel Nahodil, CSc. can be found on old website.

    Currently, we are interested in possibilities of autonomous development (design) of agent architectures based on a particular task. For unified representation of these architectures we chose framework of Hybrid Artificial Neural Networks. Projects which are under current development can be found here.

    Hybrid Artificial Neural Network Systems

    Modular systems are designed by means of connecting heterogenous nodes (subsystems) into one modular network.

    More information can be found here.

    Architect – Evolution of Hybrid Artificial Neural Network Systems

    In this research we investigate how the state-of-the-art neuro-evolution techniques can help us with automatic design of new architectures of hybrid artificial neural network systems. The aim of this research is to employ automatic design of custom hybrid architectures for a given task.

    More information can be found here.

    Nengo+Ros – Simulator of Hybrid Neural Systems

    This software tool integrates a simulator of Artificial Neural Networks (Nengo) with Robotic Operating System (ROS).

    More information can be found here.

    ROS Library for AI

    The upcoming ROS library should provide reusable sub-systems for architectures of artificial agents. The nodes provided here will be freely available and should provide implementations of selected algorithms useful in ALife domain.

    More information can be found here.