Graph memory development in a robot control architecture
Patrick M. McDowell, Cris Koutsougeras · Journal of computing sciences in colleges · 2009
The objective of this work is to develop a robot control architecture that detects and adapts to changing conditions without relying on a-priori information or human-in-the-loop calibration and setup. The focus of this work is the integration of memory based learning algorithms with coordinated observer and exploration modules through the use of a common graph based situational memory