22 - Efficient Computing in Cyber-Physical Systems [ID:4070]
50 von 517 angezeigt

The following content has been provided by the University of Erlangen-Nürnberg.

Okay, thank you very much for this kind invitation. It's actually my pleasure to be here and talk about efficient computing in cyber-physical systems.

Well, the title includes the term cyber-physical systems. What are actually cyber-physical systems?

I think the term can be traced back to Edward Lee, who in 2006 wrote that embedded systems are integrations of computation with physical processes.

So there you see already this combination of terms.

Then somewhat later at the National Science Foundation, there was a discussion about future research areas,

and people obviously discussed about extending beyond classical embedded systems, and they decided to come up with a new name,

and that new name is cyber-physical systems.

In the current description of the funding proposal there, cyber-physical systems are described as engineered systems

that are built from and depend upon the synergy of computational and physical components, so there you again have this combination.

And according to the same description, emerging cyber-physical systems will be coordinated, distributed, and connected, and must be robust and responsive.

So there you see an emphasis on the distribution and connectedness, which means that now we are talking about communication as well.

Still a little later, the German Akatek also published a report on cyber-physical systems,

and they described cyber-physical systems as network systems that are software-intensive embedded systems in a control loop,

and they provide network and distributed services, so again we see this emphasis on the communication,

because networking does not work without communication.

So obviously we have these two aspects of cyber-physical systems there within the scope of cyber-physical systems.

On one hand we have the integration with physical components, and on the other hand we have emphasis on communication.

So in order to visualize the future of cyber-physical systems, we could use a picture like this one,

where we see that everything is communicating with each other, so we might have a customer over here who is placing an order that will be communicated there to the factory,

and then the factory is linked to its providers, so we see all means of transportation are linked by communication,

and hopefully this also includes then disposal in the overall production chain.

So we see according to this picture that fabrication and logistics are obviously possible application areas of cyber-physical systems.

So if we look at the overall opportunities there for cyber-physical systems, we see a very broad scope for cyber-physical systems,

including fabrication as just mentioned, including logistics as just mentioned,

but then there are also other areas such as smart homes.

We know that for example in the future we will be able to have so-called zero energy buildings,

which means that the buildings on the average should produce as much energy as they consume,

and hopefully these buildings will also help the people who live there in the building,

so maybe we have handicapped people there, maybe we have elderly people there.

Furthermore, especially in Germany, we know that there is a transition towards a different kind of a power grid

where we have distributed production of energy and distributed consumption of energy,

and we know that only IT components can help the electrical engineers there to actually run such power grids.

Now we know that there are lots of applications there in the health sector.

We know that in hospitals for example we could use cyber-physical systems to help during surgeries there.

We also know that cyber-physical systems can help handicap people maybe at their homes,

and there are even more applications and opportunities there of cyber-physical systems.

Due to the fact that already in the term we have the term physics included,

it's very obvious that physical experiments are a special case of cyber-physical systems.

There we are linking physics to IT components.

Then I think robotics is actually one of the roots of cyber-physical systems,

one of the areas where we have already generated knowledge that's required there for cyber-physical systems.

Then we know we have lots of old buildings, some of which are not that reliable anymore.

We have falling rocks in the mountains, so therefore it does make a lot of sense to put sensors there,

and this is what actually many people are doing, and it's another case of cyber-physical systems.

For example, this is also supported by the National Science Foundation.

I've included telecommunication here as well because we have lots of real-time constraints

which are typical for such cyber-physical systems,

and we do have disaster recovery as a possible application area there for cyber-physical systems.

We know that politicians have to prepare for disasters,

Presenters

Prof. Peter Marwedel Prof. Peter Marwedel

Zugänglich über

Offener Zugang

Dauer

00:56:11 Min

Aufnahmedatum

2014-06-27

Hochgeladen am

2014-06-27 21:57:43

Sprache

de-DE

Prof. Peter Marwedel (TU Dortmund, Germany)

Computing in cyber-physical systems (CPS) has to reflect the context of the computations and, hence, has to be efficient in terms of a number of objectives. In particular, computing has to be (worst and average case) execution-time and energy efficient, while also being reliable. In this talk, we will consider optimization techniques targeting energy efficiency and worst-case execution time (WCET) minimization. 
In the first part, we will explain how the energy consumption of computing in CPS can be reduced with scratch pad memories (SPMs) and with graphic processing units (GPUs). SPMs and GPUs also help us to meet real-time constraints. We will then look at real-time constraints more closely and consider WCETs minimization. We do this by integrating compilers and WCET estimation. We will demonstrate how such an integration opens the door to WCET-reduction algorithms. For example, an algorithm for mapping frequently accessed memory objects to SPMs is able to reduce the WCET for an automotive application by about 50%. The need to seriously consider WCETs and time constraints also has an impact on applicable error correction techniques in cyber-physical systems. We will demonstrate our approach for a flexible error handling in the presence of real-time constraints which are possibly prohibiting time consuming error corrections.

Einbetten
Wordpress FAU Plugin
iFrame
Teilen