Design and Implementation of a Novel Compact Laboratory for Web-Based Multiagent System Simulation and Experimentation

This article introduces an Internet of Things (IoT)-based solution to a multiagent system (MAS) experimentation with high space utilization and high flexibility. The solution has been applied in the Networked Control System Laboratory (NCSLab), which provides hundreds of educational equipment instances, offering abundant choices for MAS experiments. Based on the networked control theory and integrated circuit technology, a novel compact architecture for online laboratories is proposed first. Combined with multiple features of NCSLab, such as online algorithm design, parameter tuning, and live video monitor, researchers can conduct interactive verification. To achieve good dynamic performance and coordinated performance in a communication-constrained network, a predictive state feedback controller for MAS is proposed. Three web-based leader–follower MAS experiments are carried out to demonstrate the effectiveness of the proposed controller and powerful configurable ability for the MAS in NCSLab. The proposed IoT-based compact laboratory provides a cost-effective and space-efficient solution for other remote laboratories.