Although the IT transformation of the 20th century appear revolutionary, a bigger change is yet to come, exemplified by the emerging Cyber-Physical Computer Systems (CPS). In these, the computer systems do not only compute abstract quantities; they are tightly integrated into physical environments with which they interact by taking sensor readings and acting on it. Such systems require a rethinking in the usual computing and networking concepts, and given that the computing entities interact with their environment, the timeliness is of increased importance. In accordance with the Moore’s law, the size and cost of nodes will continue to decrease -- thus enabling systems with ever increasing number of nodes. For example, today networks with 1000 sensor nodes have been deployed for collaborative processing of physical information. It is expected that networks with 100 000 nodes will be deployed within a few years from now. In the long-term, one can expect networks with millions of sensor nodes in operation. Such sensor systems will generate an enormous amount of sensor data and important new challenges that need to be addressed.
Applications of CPS include, among others, critical (physical) infrastructure control (electric power, water resources, gas and fuel distribution, transportation, etc.), highly dependable medical devices and systems, traffic control and safety, advanced automotive systems, process control, and manufacturing, energy conservation and environmental control, avionics, distributed robotics (tele-presence, tele-medicine) and manufacturing.
Integration of physical processes and computing is not new. Embedded systems, that is, the systems that combine physical processes with computing, have been around since long time. The revolution will come from massively networked embedded computing devices. Such extreme networking poses considerable technical challenges ranging from the (distributed) programming paradigms (languages still lacking temporal semantics, suitable concurrency models and hardware abstractions) to networking protocols with timeliness as a structuring concern, and including systems theory that combines "physical concerns" (control systems, signal processing, etc.) and "computational concerns" (complexity, schedulability, computability, etc.).
Within this initiative the CISTER/IPP-HURRAY Research Unit will be focused on research on networked-embedded sensor / actor systems, including wireless sensor and ad-hoc networks, embedded platforms (operating systems, programming paradigms, middleware), distributed embedded real-time systems and QoS-aware collaborative computing, multiprocessor/multicore systems and on applications of these computing systems to Cyber-Physical Systems (CPS) such as critical physical infrastructures or tele-physical services and applications.
Get further intuition on CISTER/IPP-HURRAY ongoing activities in the Research Skills at CISTER.