Elderly Health Promotion using Multiple Ball-robots based on Evolutionary Robot

Abstract

Eldercare is one of the most important healthcare concerns, particularly in countries whose populations are rapidly aging. Moreover, Globally the population of elderly people living alone is rising with an increasing number of people living with dementias. This paper proposes a health promotion system by using ball robot to motivate exercising for the elderly. Specifically, we use perceptual system of evolutionary robot detection based on fuzzy control. By utilizing the searching information in a limited indoor space, the robot ball can be navigated toward a target direction in real-time. In this paper, we propose a time-series of a searching method based on a steady-state genetic algorithm for visual perception. The search of the multiple robot agent is done by multi-resolution method for decreasing search space. Next, we use relative position distance and radius of the ball robot as inputs to the fuzzy controller. Therefore, we get the output value, speed and direction angle, as robot’s moving command of next state. Experimental results of boccia exercise also show the effectiveness and the practicality of the proposed robot system for typical healthcare services.