Environmental feature extraction and mergence: make the past serve the present

Abstract

This paper proposes a connectionist model to learn a spatial representation of the world based on temporal memory of perceptions and actions of a mobile robot. It is constructed at run-time to merge past experiences and retrieved in later runs to guide the robot to perform the navigation task. A coding strategy is introduced to extract the directional information from the perception sequence, which endows the robot with localization ability. The temporal sequence processing network (TSPN) transforms routing knowledge learned from robot's experiences into temporal characteristics of cell firing and enables the implicit building of a world representation. The navigation system integrating TSPN and a reactive safeguard module performs collision-free navigation, dynamic landmark and heading detection, route learning and path planning in a noisy world. The simulation and real world experiments demonstrate the flexibility and robustness of the system.