Design of autonomous navigation system based on affective cognitive learning and decision-making

Abstract

A new autonomous navigation control system is presented for mobile robots based on the affective cognitive learning and decision making (ACLDM) model. The behaviors of robot navigation are designed by dynamic system approach, which has a sound theoretical foundation for the system stability analysis. Cognitive states for work environment of the mobile robot are gotten from a pattern classifier based on Adaptive Resonance Theory-2 (ART-2) network. Then rational strategies for behaviors coordination are developed by on-line affective cognitive learning. This control strategy can make the mobile robot navigate autonomously in unknown environment. The designed behaviors can guarantee that the robot navigates safely by choosing an appropriate velocity. Simulation studies have demonstrated that the integration of the affective system with cognitive system can speed up the learning process, and the proposed strategy can effectively improve the capability of robot's autonomous navigation.