Master of Science in Embedded Computing Systems

Course on Real-Time Embedded Systems

Offered by
University of Pisa (Department of Information Engineering)

Why Embedded Systems?

Over 98% of the computers existing in the world are embedded, that is, integrated into a larger system with the purpose of managing its resources and monitoring/controlling its functions using special hardware devices.

In the last thirty years, embedded computing systems grew exponentially in areas like industrial automation, avionics, automotive, telecommunications, consumer electronics, and robotics. The number of application domains is rapidly expanding in other fields, like sport and medicine (portable monitoring devices), agriculture, game and toy industry, civil protection, and intelligent transportation systems.

How embedded systems differ from other computing systems?

General purpose computers (like PCs or tablets) are developed to provide high average performance to user applications, without paying attention to worst-case delays, which in certain rare and unpredictable circumstances could be very large. In other words, general purpose computers trade performance with predictability. While sporadic long delays are normally tolerated by humans, they could cause a critical failure in a computer controlled system, like a robot or an aircraft.

To avoid such a problem, embedded computing systems must be designed to guarantee bounded worst-case response times, rather than good average performance. Unfortunately, providing such a timing guarantee is not trivial and requires several mechanisms, techniques, methodologies and tools that are subject to a continuous evolution, due to the progress of computing platforms.

Objectives of the course

Unfortunately, a correct control algorithm is not sufficient to ensure a correct behavior of the controlled system, because unbounded execution delays due to inter-task interference, may compromise the performance as well as the stability of the system. To address this problem, this course will show

How to model computations to take time constraints into account when controlling a system.
How to derive timing constraints from the application.
How to combine control and programming.
How to design and analyze software to make sure it behaves as you expect.
How to analyze timing properties.
How to manage overload situations.
How develop real-time software for safety-critical systems.
How to implement discrete control systems on a real-time operating system.
How to simulate physical systems described by differential equations.
How to simulate sensors & actuators to create virtual environments useful for experiments.

In general the course is aimed at providing both theoretical foundations and practical aspects in the topics listed above, forming engineers capable of designing, analyzing, implementing, and testing complex computing systems using the most advanced methodologies.

Strong international contacts

The teacher of this course has strong international collaborations with the top research teams in the world and has established contacts with the major industries in the filed. Examples are listed below.

Universities

University	Location	Topic
University of California at Berkeley	Berkeley, California, USA	Hardware/Software Co-design
University of Illinois at Urbana-Champaign	Urbana-Champaign, Illinois, USA	Embedded systems architectures
University of Virginia	Charlottesville, Virginia, USA	Real-time sensor networks
University of Pennsylvania	Philadelphia, Pennsylvania, USA	Robotics and real-time control
University of North Carolina at Chapel Hill	Chapel Hill, North Carolina, USA	Real-time scheduling theory
University of Lund	Lund, Sweden	Real-time control
Uppsala University	Uppsala, Sweden	Formal methods for software verification
Malardalen University	Vasteras, Sweden	Real-time systems
Royal Institute of Technology (KTH)	Stockholm, Sweden	Real-time control
University of York	York, UK	Real-time systems and multi-core platforms
University of Porto	Porto, Portugal	Real-Time Networks
Polytechnic Institute of Porto	Porto, Portugal	Distributed Embedded Systems
Technical University of Vienna	Vienna, Austria	Timing analysis
Technical University of Kaiserslautern	Kaiserslautern, Germany	Real-Time systems
Technical University of Munich	Munich, Germany	Real-Time Embedded systems
University of Lille	Lille, France	Real-time systems
University Carols III of Madrid	Madrid, Spain	Multimedia systems
Polytechnic University of Madrid	Madrid, Spain	Multimedia systems and QoS management
University of Catalonia	Barcelona, Spain	Real-time control
University of Cantabria	Santander, Spain	Real-time distributed systems
University of Valencia	Valencia, Spain	Real-time operating systems and hypervisors

Large companies

Company	Location	Topic
Ericsson	Lund, Sweden	Mobile communication systems
Philips	Eindhoven, The Netherlands	Multimedia systems
Airbus	France	Safety-critical software for avionics
Bosch	Germany	Engine control for automotive systems
Magneti Marelli	Italy	Engine control for automotive systems
Huawei Research Center	Pisa, Italy	Automotive systems
Hitachi Rail STS	Napoli, Italy	Railway localization systems
Wind Rivers	Torino, Italy	Real-time operating systems
Tecnalia	Bilbao, Spain	Automotive systems
Gambro	Modena, Italy	Dialysis Machines

Small-Medium Enterprises

Company	Location	Topic
Accelerat	Pisa, Italy	Hypervisors and hardware acceleration
Evidence	Pisa, Italy	Real-time operating systems and software
IDS Georadars	Pisa, Italy	Georadars
Aitronik	San Giuliano Terme, Pisa, Italy	Drones and autonomous vehicles
ALES	Roma, Italy	Embedded system design
Absint	Germany	Timing analysis tools
Rapita Software	York, UK	Timing analysis tools
Timing-Architects	Regensburg, Germany	Simulation tools for automotive systems
Seabear Diving Technology	Graz, Austria	Diving systems
TimeSys	Pittsburgh, Pennsylvania, USA	Real-time kernels and tools

Possibility of international theses

Master theses in collaboration with industries or foreign universities are possible and highly encouraged.