Place Recognition and Topological Map Learning in a Virtual Cognitive Robot

An ACT-R cognitive model is used to control the spatial behavior of a virtual robot that is embedded in a three-dimensional virtual environment, implemented using the Unity game engine. The environment features a simple maze that the robot is required to navigate. Communication between ACT-R and Unity is established using a networkbased inter-operability framework. The ability of the robot to learn about the spatial structure of its environment and navigate to designated goal locations serves as a test of the ability of the framework to support the integrative use of cognitive architectures and virtual environments in a range of research and development contexts.