Human-Centered Dynamic Service Scheduling Approach in Multi-Agent Environments

Abstract

As robots become more versatile and combined with a variety of Internet-of-Things technologies, they will be able to serve humans in their daily environments. To provide services by satisfying various human requests, several robots must take turns performing a series of tasks that constitute the service. Because the order of service delivery may differ according to user requests, sequential interdependencies between tasks should be considered. Therefore, we propose a dynamic service scheduler consisting of dynamic sequencing and allocation that can handle scheduling of tasks with user requests such as prioritizing certain tasks or actively changing their order in a multi-agent environment. We experimented with the proposed method in four situation scenarios by building a virtual reality smart office consisting of multiple robots with a robot arm, mobile robot, and smart lamp. The results demonstrated the feasibility and effectiveness of the proposed approach by satisfying the user requirements in different situations. The proposed approach constitutes a basis for further development of efficient in-office and at-home multi-agent environments.