morris The goal of the MOdular Neural Exploring Traveling Agent (MoNETA) project is to develop an animat that can intelligently interact and learn to navigate a virtual world making decisions aimed at increasing rewards while avoiding danger. The animat, which is a virtual agent living in a virtual environment, is designed to be modular: a whole brain system, initially including fairly simple modules, will be progressively refined with more complex and adaptive modules, and will be tested in increasingly more challenging environment. The animat brain is designed in Cog Ex Machina (Cog), the software realized by HP in collaboration

with Boston University in the DARPA SyNAPSE project. Cog, which can run on CPUs, GPUs, and will run on memristive-based devices, allows to pack large-scale, highly interconnected, plastic, heterogeneous neural models that make up the animat brain in a low-power, high density chip which is suitable for implementing portable petascale neural-based computing. The current plan is for the animat to replicate a classic rat experiment, the Morris Water Maze (left), by Fubruary 2011, and progressively simulate more complex mazes used in rat experiments. Further evolution of the model will use the Iterative Evolution of Models (ItEM) project software.

Below a video of MoNETA as it swims in the VIrtual Morris Water maze (November 26, 2010), reaching the goal (the submerged platform) autonomously learned by combining visual, emotional, and world-based information.

The animat currently "lives" in a Virtual Environment world (see below).


MoNETA includes the following macromodules:


  1. Sensory Object Recognition and Tracking (SORT, in pink): Currently. SORT includes a fairly simple what and where system. The where system was also tested in a robotics platform to show its use in realistic environments. Below, a video of a few submodules of the SORT system as the animat navigates in the virtual Morris water maze.
  2. sort
  3. Spatial Planning and Allocentric Representation of Knowledge (SPARK, in green): the current MoNETA brain implements navigation via a neurally-based bidirectional graph search for route planning to learn locations of rewards/punishments, and determine the next steps for the animat within the virtual environment.  
  4. MOtivation, Reward, and goal SELection (MoRSel, in yellow): MoNETA implemets a simple drives representation that, in conjuction with the rest of the animat brain, contributes to generate "intention" in the simulated agent.
  5. Below, a video of SPARK and MoRSel as the animat navigates in the virtual environment.
  6. moneta

The two videos below show MoNETA performing Morris water maze task with the unknown goal. Ohe left, there the first run is the rat before it learns the location of the submerged platform (video at16x speed), on the right a run showing the animat behavior with the goal known from previous runs.



Below is The Neuromorphics Lab VIsually GUided Adaptive Robot (VIGUAR), a step towards the implementation of complex memristive-based MoNETA brains in mobile robotics platform.