Affective-Model Based High-Level Controller for Human-Robot Applications

AbstractThe paper presents a real-time affective state detection namely the engagement level detection by using fuzzy classifier that can be applied to human-robot interaction. In order to develop the fuzzy classifier, the engagement model is developed using the data collected from a controlled design experiment. In the experiment, the data collected are from the total number of endogenous eye blinks and the total error from the trajectory the subjects have to follow in completing specific tasks. For the tasks, the subjects are asked to track a set of prescribed paths within the allocated times and have to obey different speed constraint. Various shapes of trajectories are given to the subjects in order to study the level of engagement while performing the task. The data then are used to develop the fuzzy classifier to measure the level of engagement (LOE) of the subjects. Following the experiments, a series of questionnaires are given to the subjects to validate the engagement model developed. The result from the fuzzy classifier is applied on a robotic model that has linear motion featured control. The LOE can be used to adapt the speed of the robotic platform model which is useful for the human-robot interaction. In this paper, the engagement model is in the form of fuzzy classifier is designed as a higher level controller using the discrete-event system (DES) approach to control the speed of the robotic platform. Preliminary analysis on the high-level controller shows a promising result for future research in application for robot-assisted rehabilitation