This paper offers an innovative system to manage Internet of Things (IoT) and Artificial Intelligence (AI) platforms based on Human-in-the-Loop paradigms. Our solution is implemented and validated in a testbed that was also built to the monitorization of the surroundings of students in favor of improving their academic performances by using a Human-in-the-Loop AI system. Therefore, the system is composed of a lot of different sensors that are incorporated into different physical devices. With this in mind, we explore the concept of management of an IoT network of these heterogeneous devices and evaluate the path to follow, considering the components that already exist in the system. We use the Lightweight Machine-to-Machine (LwM2M) as a management protocol, even though it is recent and there was not a lot of information on its use in cyber-physical systems. To integrate this protocol into this innovative platform, we present two approaches. The first one is based on the Eclipse's Leshan project and the second works with FIWARE's LwM2M IoT Agent. Concerns about the energy consumption by the Raspberry Pi and the smartphone application with the protocol were raised, so performance tests were made, and the results were favorable. The percentage of traffic introduced by LwM2M was also obtained.
Keywords
#human-in-the-loop AI
#energy efficiency
#Leshan
#cyber-physical systems
#heterogeneous IoT devices
#student monitoring
#academic performance
#Raspberry Pi
#mobile application
#network traffic
#Internet of Things
#Computer science
#Human-in-the-loop
#Loop (graph theory)
#Energy management
Track outputs, projects, and reports in one place.
Sign in to update your profile, manage outputs, and share results with your institution.
Login
Enter your credentials to access your workspace.
Access to the portal is created by your entity's administration (ISLA GAIA or CEOS.PP). If you have not received the invitation email yet, please contact them.
Recover Password
Enter your institutional email to receive a reset link.