Development of a Cloud-based Computational Framework for an Empathetic Robot

This article presents the development and preliminary evaluation of an empathy controlled robot. Such a robot is one step forward towards industry 5.0, as it provides a theoretical framework to enable the performance of the robot to be customized to suit the needs of both the task as well as the operator. An inventive step is taken through the separation of computational resources based on whether the algorithms are addressing functional or experiential needs. The paper therefore addresses the requirement for new approaches that can be employed in the design of mobile robots to reduce cost, power consumption, and computational burden of the system. We propose that tasks requiring real-time and safety critical control are processed using dedicated on-board computers, whereas functionality dedicated to system optimization, machine learning and customization are handled through use a cloud-based platform. In this paper, key components of the architecture are defined, and the development and preliminary evaluation of an exemplar robot capable of changing its behavior in accordance with the perceived emotional state of an operator's voice is presented.