Stacks Image 0

The ExpCog Framework

ExpCog is a high-level cognitive robotics framework aimed at integrating logic-based and cognitively-driven agent-control approaches, qualitative models of space and the ability to apply these in the form of planning, explanation and simulation in a wide-range of robotic-control platforms and simulation environments. In addition to its primary experimental function, the framework also has a utility toward didactic purposes, e.g., as a teaching and experimental aid in courses on Artificial Intelligence, and other specialized courses, tutorials involving some form of high-level robot control.
Stacks Image 1

Reasoning - Control

The primary aim of the framework is to provide a suite of different control approaches that may be used for representing and reasoning about dynamic environments.

The suite of control approaches available with the framework also constitutes the most important (functional) component of the overalll experimental framework. It consists of a collection of several different formal techniques, both logic-based and cognitively-driven models, that can be used as control mechanisms in robotic domains or be used to reason about changing spatial environments in general.

Initial Scope:

  • STRIPS
  • Belief-Desire-Intention
  • Situation Calculus
    • Golog
    • Indigolog
  • Event Calculus
    • The DEC Reasoner
  • Fluent Calculus
    • FLUX
  • Answer Set Programming
    • HPX (in oclingo)

Robot Platform Independence

The framework is independent of any particular robotic system/platform or agent simulation environment thereby ensuring applicability in a wide-range of real or simulated environments. An adequate level of abstraction between the high-level reasoning component and robot hardware / simulation platform is implemented by a Controller Communication Interface (CCI). The CCI provides the abstraction between robotic or simulation platforms and the experimental framework.

Presently supported:

  • Gazebo
  • ROS - Robot Operating System
  • Robotcub - iCub
  • v-rep
  • Rolland Wheelchair
Broadly, Project ExpCog is intended to be a long-term initiative aimed at performing an integrative role involving research from robotics, reasoning about action and change, spatial and temporal reasoning, and the robotics centred Artificial Intelligence community at large.
CoSy
SFB/TR 8 Spatial Cognition
University of Bremen