Face Memorization Using AIM Model for Mobile Robot and Its Application to Name Calling Function

• Main Authors: Masahiko Mikawa, Haolin Chen, Makoto Fujisawa
• Format: Article
• Language: English
• Published: MDPI AG 2020-11-01
• Series: Sensors
• Subjects: social human–robot interaction; face memorization; face classification; user acceptance; AIM model; sleep function
• Online Access: https://www.mdpi.com/1424-8220/20/22/6629

We are developing a social mobile robot that has a name calling function using a face memorization system. It is said that it is an important function for a social robot to call to a person by her/his name, and the name calling can make a friendly impression of the robot on her/him. Our face memorization system has the following features: (1) When the robot detects a stranger, it stores her/his face images and name after getting her/his permission. (2) The robot can call to a person whose face it has memorized by her/his name. (3) The robot system has a sleep–wake function, and a face classifier is re-trained in a REM sleep state, or execution frequencies of information processes are reduced when it has nothing to do, for example, when there is no person around the robot. In this paper, we confirmed the performance of these functions and conducted an experiment to evaluate the impression of the name calling function with research participants. The experimental results revealed the validity and effectiveness of the proposed face memorization system.