Presentation: Applied CI/CD: Enabling Creativity @Volvo Trucks

Location:

Duration

Duration: 
1:40pm - 2:30pm

Day of week:

Key Takeaways

  • Hear a completely different use case of technology that you don’t often hear at software conferences.
  • Learn how the automotive industry (particularly in the space of Big Trucks) embrace the tools and technology we as software developer’s take for grant in innovative ways.
  • Understand how virtualization is changing the CI/CD pipeline for complex trucking systems..

Abstract

Enabling Creativity – the truck system with Continuous Integration and Virtualization demonstrated.

The automotive industry is on the verge of a fundamental revolution as the vehicle themselves becomes autonomous – Continuous Integration and Continuous Delivery using virtualizations linked to real trucks will play key roles in making this happen. The heavy duty truck industry will be in the center of this major change. Trucks are already today advanced technical products with internet integrated functions; however, this is still very limited compared to what we will see in the upcoming years. This change will offer fascinating possibilities for innovation and creativity when trucks will be parts of integrated autonomous transport solutions.

Interview

Question: 
CCon: What’s the motivation for your talk?
Answer: 
Peter: I used to work with mobile phones. Mobile phone has around 14 sensors. Trucks today, without all this new technology, have almost a thousand sensors on them. So if we call a device with 14 sensors on it a smartphone, then I definitely think that we have smart trucks today. You have seen them a lot of times when you drive around, but I guess you haven’t thought about how complex they are. I compare truck development time with the time it took Google to become Google: we spent the same time building trucks as it took Google to become Google. That obviously tells us something that we need to learn from companies like Google They have the speed and the creativity that we haven’t been able to match at all. So in this talk I’m talking about using creativity to build smart trucks.
Question: 
QCon: We are seeing more and more IT companies entering the automotive industry. How is IT affecting the automotive space?
Answer: 
Peter: It’s fundamentally changing it. If you look at the truck of the future, you will find more or less a big computer. The truck will likely leverage virtualization (things like VMware or something similar). It will run Java or C/C++. The computational power will be 100’s and 100’s times bigger than today’s cars and trucks. The truck is completely linked with the Internet (as it is today), but it will be more an autonomous vehicle. Exactly when that will happen is obviously a big question, but it is just a matter of time. These trucks will probably be one of the most complex technical things that we see in everyday life.t will be a quite fascinating thing to see.
Question: 
QCon: So you said these automobiles and trucks are going to be running virtualized containers, with virtual machines running in them. What is the use case VM’s in a big truck?
Answer: 
Peter: It is extremely powerful because if you look at a modern truck (or a modern car), you have huge computer systems in them. Let’s say, 40 computer systems in a truck (in our trucks, we have 36 networks). So we have many, many thousands of signals on those networks and that’s with today’s technology. If we were to scale that and get autonomous driving vehicles going, there is no way that that will scale without completely changing basic technologies. Virtualization has the capability of running several computer systems (like brake, or engine systems simultaneously). The use of virtualization allows the possibility of isolation and managing versions across these systems.. This is critical t because to test such a system requires a lot of road mileage that just isn’t possible without the idea of virtualization.
Question: 
QCon: How do you leverage virtualization with these trucks?
Answer: 
Peter: A virtual truck is the code that runs on the actual truck but running in a virtual environment. A virtual truck has all the systems modeled with the code that makes it possible to leave the actual trucks behind. Everything is virtualized. It is a complete truck. It brakes, and it does all the things that a real truck does. But through virtualization, it becomes possible for each and every developer to have their own their own test fleets of trucks (at a very low cost).
The scale of a truck project is enormous. It is thousands and thousands of engineers building a truck. Most of them are actually not dealing with the software, because the software is not the most complex part. The most complex part is handling the environment. It’s the roads, environment temperatures, roadway systems and all of those things. If you were to test each of these physically (which today is still possible) it would take a long time. You would then either need a complete truck (or a hardware based simulation costing say €1 million), so we could do maybe 20 tests an hour. With something like Docker, we have the capability to produce the same truck virtually. Now we can have individual trucks and test those trucks in the Cloud. So it is critical for us to leverage virtualization.
Question: 
QCon: What are the technology challenges for the auto industry today?
Answer: 
Peter: The challenge in today’s heavy industry is that the round trip from when you make a change in your code until you know if it works on the roads with customers could be 30 to 40 weeks. That is slow if you compare to the construction of a website. It is definitely needed to have very, very short iterations. The technology that we are trying to build here is enabling substantially shortened times between when you check code in until you know if it works in the truck. If we were able to reduce that time, we could give you (as a developer) the time to be a developer instead of just waiting months for the feedback.
Question: 
QCon: Are we close to having autonomous capabilities in our vehicles?
Answer: 
Peter: I think it is quite interesting to see that the hype is so much around cars and not with trucks. Trucks are exciting to drive, but their main focus is transporting goods. That means that there is less resistance. So acceptance to autonomous vehicles in heavy industry is much higher. Also, in some places like mines and forests it’s really dangerous, so it’s very good if you don’t have drivers in them. It’s even possible today without any change of legislation to drive automated trucks in mines because they are private property. The real opportunity is in the heavy vehicle industry because the business case is much more obvious.
Question: 
QCon: What is the main thing you want people to leave with from your talk?
Answer: 
Peter: I think they should really feel that compared to the changes we have seen, for example, in the mobile industry, the change we are going to see in heavy automotive is bigger. They should feel that it’s something close to them as well because of things like virtualization. They should feel that there is a tremendous opportunity in this space. to work with heavy trucks weighing 76 tons with engines of up to 750 horsepower. I hope they learn that these are huge engines, but, in the end, it’s the same technology of things we learn to do in your smartphone app. These things the audience are experts with is completely useable in the trucking industry That’s really exciting.

Tracks

Covering innovative topics

Monday, 7 March

Tuesday, 8 March

Wednesday, 9 March

Conference for Professional Software Developers