An Online Hierarchical Fuzzy Rule Based System for Mobile Robot Controllers

摘要

The introduction of automated robots has revolutionised the manufacturing industry. The further development of autonomous mobile robots capable of functioning in unstructured and dynamic environments is highly desirable. This paper outlines a novel method for the online development of an interpretable mobile robot controller using supervised learning. An information theoretic approach is used to control the rate of expansion in a Hierarchical Fuzzy Rule Based System (FRBS). Experimental results, on a simulated mobile robot, are provided to demonstrate how the uncertainty tolerated can be used to control the trade-off between accuracy and interpretability.