|
| Re: IEC 61499 Programming Standard: Will it Become Mainstream?: [message #1840467 is a reply to message #1840459] |
Thu, 15 April 2021 17:04   |
Eclipse User |
|
|
|
Thank you Bill,
Your article and discussion with Alois is enlightening. I am currently finishing my doctoral studies, which have focused on fault-finding for IEC 61499 systems using multi-agent approaches. During that time, I have worked with a number of different IEC 61499 development platforms.
For me, the one thing that has stood out is the depth of the 4diac/FORTE community. Alois and his team put so much time into supporting 4diac users. That is genuinely appreciated around here. That makes a huge difference to those of us who are working out how to design and deploy IoT systems that are robust and can be verified formally. I came previously from a world of development for automation where hacking together C/C++ code with little rigor and verification predominated. Working in IEC 61499 using 4diac is demanding, but I am learning how to deliver robust applications that I believe will stand up to the demands of safety-critical certification. That's the next area beyond my doctoral work that I want to head into; large IEC 61499 systems and formal safety evaluation methods. I think that being able to meet standards such as DO-178C for the aerospace industry is challenging for IEC 61499, but the framework has tremendous potential to address the obstacles involved. In my AUT SERL laboratory, we have an emphasis on formal methods and IoT security. Security for IEC 61499 should be a primary focus for many academic and industry practitioners going forward. From what I observe here, I think 4diac has a road map which is taking these issues seriously.
Unfortunately, that level of support is not evident for the other IEC 61499 development platforms and in other communities. In the long term, I think it will be community support and collaboration that is going to drive the adoption of IEC 61499. So thank you Alois, your team, and the 4diac community - I think things around here are definitely going in the right direction.
I look forward to talking with you further Bill.
Kind regards,
Badger Dowdeswell.
https://academics.aut.ac.nz/barry.dowdeswell
[Updated on: Fri, 16 April 2021 01:26] by Moderator |
|
|
|
|
| Re: IEC 61499 Programming Standard: Will it Become Mainstream?: [message #1840627 is a reply to message #1840545] |
Tue, 20 April 2021 23:26 |
| Eclipse User |
|
|
|
Hi Alois,
I also like the nickname "Prof AZ" too.. but we must be careful or else it will catch on with your students :)
Thank you for asking for my opinion on this. C++ has some tremendous advantages over Structured Text in environments like 4diac/FORTE if engineers are willing to take the time to learn it. You will see in recent posts from me that I have been interfacing sensors to REAL hardware as part of the final experimental phase of my doctoral research. However, the driver code for the temperature sensors such as the DS18B20 I am using could not be written easily in Structure Text (at least I could not see how to do it ... ).
In hindsight, learning C++ under C:++11 was so much easier than I expected since many of the constructs are much closer to all my Java experience; C has come a long way by the time we got to C++11 in 2021 !!!! Also, learning C++ allowed me to be much more confident about diving down into the core of FORTE. That taught me so much about how it worked. Sometimes when I was stuck trying to code something, I would run a search on the core library to find a function that I could learn from. That's how I found out how to print CIEC_STRING variables out with cout : there is a getValue() method on CIEC_STRING that I had not found before. That means when debugging, I can compile in all the console messages I need. While the FBTester is a great tool, it is not enough when trying to debug a complex set of interactions across multiple blocks. Real embedded programmers have to work down on the bare metal some times !!!!
Also, I can control casts properly to ensure that in some of the complex maths we are doing inside events, we get the correct precision. Again, that could be problematic in Structured Text.
During my Post-doctoral that starts in a few months time, I intend to start writing some tutorial guides for 4diac/FORTE including using C++ routines . Professor Roopak Sinha and I have discussed this as an option for me. We feel it would be useful for both our students here at AUT that we want to introduce 4diac to, and for the 4diac community at large. Also, I am interested in documenting more examples of getting sensors and actuators going on platforms such as Raspberry Pi and the WAGOs we have in our AUT laboratory. Again, all this is just a little part that helps to foster the collaboration that you and your team have facilitated Prof AZ.
Cheers,
Barry
|
|
|
| Re: IEC 61499 Programming Standard: Will it Become Mainstream?: [message #1840671 is a reply to message #1840627] |
Wed, 21 April 2021 15:53 |
| Eclipse User |
|
|
|
Hi Barry,
thanks for you long an insightful post. Yes I agree that in several situations currently it is still necessary to debug 4diac FORTE or your FBs in C++. I really hope that this will not be needed in the future. There are some activities in that direction. We will keep the community updated.
Also there will be quite some improvements in our overall ST support. Hopefully we already have something for the autumn release.
Cheers,
Alois
|
|
|
Powered by
FUDForum. Page generated in 0.03350 seconds
Home