Urban systems like traffic, energy, and outdoor lighting are managed by self-contained embedded systems ? though the managed processes are deeply interconnected. New applications and collective optimization require integration of these systems which represents a truly ?systems of systems? integration problem: these urban systems evolve independently, have their own purpose and internal policies which must not be affected by such integration and have their own management. As a result, the integration hasto manage emergent behavior and take non-availability of components as the norm rather than the exception. This is particularly challenging when control loop span across several systems. ACCUS aims at three innovations: 1. Provide an integration and coordination platform for urban systems to build applications across urban systems 2. Provide an adaptive and cooperative control architecture and corresponding algorithms for urban subsystems in order to optimize their combined performance 3. Provide general methodologies and tools for creating real-time collaborative applications for systems of systems ACCUS studies a set of so-called converged scenarios that span across urban systems to investigate requirements and defines a reference architecture for the integration of urban systems, based on semantic descriptions. Simulations with software and systems in the loop will be supported. The entire development and integration will be supported by tools. The ACCUS system and tools will be validated and demonstrated in two extended use cases in Poland and Italy and with at least four different urban systems. The input of ACCUS is the work of several other ARTEMIS projects that focus on the self-contained urban systems themselves. ACCUS goals are ambitious.In order to achieve these goals a well-established and high-quality consortium with a broad scope of complementary disciplines and a long experience in European and National programs is lined up to do the job.

Keywords
System of systems, integration, adaptive control, real-time collaborative applications