Construction and evaluation of a decision-making system for autonomous cars based on a decision table

Abstract

We propose a decision-making system for autonomous cars based on a decision table. By using the decision table, the proposed system offers faster reasoning speed than a conventional system using ontologies and reasoning rules. Furthermore, we propose an automatic conversion system from a set of ontologies and reasoning rules to a decision table, thus, the result of reasoning is equivalent to the conventional one. Experimental results show that the proposed system realizes approximately 40 times faster processing time than the conventional system.