MAIS-S: Multiagent Intelligent Surveillance System

Start date: 2010 Expected completion date: 2013
PIs: Francisco António Chaves Saraiva de Melo (INESC ID / IST/UTL) and Manuela M. Veloso (CMU)
Team: INESC ID, Instituto Superior Técnico/Universidade Técnica de Lisboa, ISR, Carnegie Mellon University (CMU)
Company: Observit, Lda

Keywords: Planning in Multiagent Systems; Sequential decision-making under uncertainty; Sparse interaction; Intelligent surveillance systems

With the generalized use of "intelligent technology", the interaction between multiple smart devices poses interesting challenges both in terms of engineering and research. One interesting aspect of this phenomenon in the context of this project is the appearance of networks of heterogeneous devices that must operate in a fully distributed manner while sharing information necessary to complete some pre-assigned task. In this project, we propose that such complex networks be modeled as multiagent systems where each node corresponds to an agent. This interpretation suggests several interesting research avenues, some of which will be the focus of this project. We propose the use of a class of decision-theoretic models - Dec-POMDPs and specializations thereof - that naturally captures the decentralized nature of these networks in terms of local perception (the information that each node can acquire per se), interaction/communication (the exchange of information between the nodes) and local actuation (each node processes locally the available information and acts accordingly). In fact, one can argue that several systems typically considered in a fully centralized fashion (such as surveillance systems) could potentially benefit from this multiagent view of the network.

We are interested in heterogeneous surveillance networks that include different kinds of nodes, possibly with different perceptual and actuation capabilities, as well as different processing power. From this perspective it is natural, for example, to have nodes corresponding to cameras mounted on mobile robots. The control and even the positioning of the robots should be decided locally but in such a way to globally optimize the performance of the network. The proposed work will follow one of two main lines. We will formalize/abstract several fundamental problems typically found in most surveillance systems as optimal decision-making problems (Tasks 1-2). Within this formalism we will tackle these problems in a principled way and propose solutions that offer some theoretical guarantees of performance. On the other hand, we will investigate several practical problems faced in deploying such a heterogeneous network (Tasks 3 and 4) and bring the results from our research in Tasks 1-2 into practical use.

Articles published in the Portuguese media:
   The MAIS-S Project at the Portuguese Television SIC Notícias(June, 2013)
   Follow the Robot (May, 2013)
   Project Develops a Specific Type of Intelligent Surveillance [Ciência Hoje, Diário dos Açores, Mundo Português, August 2011]