A multi-agent architecture for mobile sensing systems

Abstract

Mobile sensing systems based on smartphones, connected vehicles and integrated sensors on new mobile devices have become an important alternative for the development of intelligent services in large urban environments. Massive data collection and its real-time analysis are essential for big cities to move towards energy efficiency, sustainable mobility, protection of the environment and economic sustainability. Current research and applications are mainly focused on the use of individual devices and the analysis of information on a single domain (e.g. activity recognition). However, it is still necessary to provide solutions for social problems based on smart mobile devices connected to the city. In this paper, we present an architecture for mobile sensing systems in large cities based on the intelligent agent paradigm and multi-agent systems. The presented platform provides support for multi-purpose machine learning services, implementing expert learning agents in each domain where the system collects data. Furthermore, the main challenges in mobile sensing systems such as scalability in crowded environments, handling of a large amount of data and the increasing appearance of sensing devices are addressed by the architecture due to the agent paradigm and multi-agent systems suit these demands naturally.