Mirob: an intelligent service robot that learns from interactive discussions while handling uncertain information in user instructions

Abstract

This paper presents about an intelligent service robot named Moratuwa Intelligent Robot (MIRob) that can acquire knowledge through interactive discussion with the user while handling the uncertain information in the user instructions. In order to facilitate this behavior, a finite state intention module has been introduced to manage the interaction between the user and the robot. A set of states has been defined to acquire and update the knowledge as well as in order to perform the actions required to satisfy the user instructions. Dialogue flows and patterns have been defined for each state in order to maintain the interaction with the user. The defined dialogue flows and the dialogue patterns are presented. Furthermore, the concept has been developed in such a way that the system is capable of updating a Robot Experience Model (REM) according to the acquired knowledge. In order to evaluate the performance of the system, experiments have been carried out in an artificially created domestic environment. The capabilities of the proposed system have been demonstrated and validated from the experimental results.