Enabling Virtual Proof of Validation for Autonomous Driving — on Tezos
We sit down with Carlo van Driesten from BMW on what he’s building for the Tezos Ecosystem
For quite some time within the Tezos community, we’ve been aware of a very ambitious and innovative effort focused on providing virtual proof of validation of autonomous driving functions — on Tezos. The subject is a complex one and to shed more light into the work being done within the ENVITED ecosystem, Carlo van Driesten of BMW Group was kind enough to help extrapolate just exactly what is being worked on.
What is ENVITED?
The ENVITED ecosystem comprises of several industry leading automotive manufacturers such as BMW, Audi, Porsche, Daimler, and many other notable partners. Born from the Automotive Solution Center for Simulation e.V., the ENVITED ecosystem work group is a collective focused on defining a methodology for standard-based and machine-based proof of validation for highly autonomous driving (HAD).
ENVITED is a distributed ecosystem with several players and moving parts all interconnected under a common settlement layer.
Automated driving functions must comply with regulatory requirements. Standards are an important basis to ensure quality which automobile manufacturers must adhere to. The correct implementation and use of standardized data and components are proven by certification — which requires verification by the appropriate regulatory entities and test organizations.
Numerous pieces of information and probabilities must be accounted for in the design of AD functions. The development of the standards behind them will have to make use of a virtual setting as — with so many pieces of information/probabilities to be considered, using a public environment will be next to impossible to account for each potential event that may arise in a real-world setting.
So, within this virtual world — the same things done within the physical world with assembling a car must also be replicated within a virtual setting as well. Thus, having a single automobile manufacturer develop their own AD functions will likely result in some inconsistencies and errors that could prove fatal in the real-world.
So, rather than having an automotive manufacturer build everything “solo,” within the ENVITED ecosystem — several large automobile manufacturers and other players are all sharing a virtual testing environment, testing results, data, designs, and advancing at a greater pace towards achieving certification and providing proof of validation for AD functions.
Benefits of Blockchain Technology
In order to prove validation of specific AD functions, some challenges such as immutable documentation, transparency, and automated contract execution arise.
Within blockchain, one of the key benefits of blockchain technology is immutable record storage and retrieval thereof. In relation to the ENVITED ecosystem, blockchain can be utilized through the back end of each automobile manufacturer and entity (since they are all connected together under a settlement layer) to share development processes.
Ultimately, the usage of blockchain technology (in this case Tezos) will allow for indisputable proof of validation of a car having X automated firmware, and traceability to show that required testing and processes were followed.
As a counter-argument of using blockchain, and rather using digital signatures instead — Carlo explained why that may not be the most efficient way for proving validation.
In this case, the magical word is, in my opinion — revocable certificate. So, take a highway example for instance. If you want to create a test area, you need a map of that highway and someone must drive along, measure that map, and create it. The property of a map is, that it will be outdated at some time. Because, you have some construction work, or you have to be sure again that the map is actually still as you thought it was.
So, the validity of a map can be certified — but it can only be certified for a certain amount of time. So, if you think about something like an SSL certificate, which you give to someone. As a security measure, they have an expiration date — for example, one year. In this case, you could say I’m giving this map a restricted time validity of one year and the problem might be solved. But, you have a lot more flexibility if you say this map is valid until I say it’s not valid anymore and I’ve revoked that certificate. So, blockchain is very powerful if you have the use case of revocable certificates.
To further understand the decision basis behind choosing to utilize Tezos within the ENVITED ecosystem, I decided to ask Carlo van Driesten a few questions surrounding his decision-making basis for using Tezos.
What was the decision basis behind choosing to utilize Tezos?
I thought there were many reasons. As a company person, I know that processes are something that are very powerful and necessary. So, Tezos has a working change management process for the ratification of off chain collaboration efforts. It’s something so powerful and so important — that this is the main reason for me to propose Tezos as a possible technology.
Because, I know there is a way of upgrading something that is decentralized and distributed. Additionally, that there is a way of having influence on that system along our own development process, be able to publish our requirements and observe if the technology is moving
into the needed direction.
I think the Tezos community is a community that actually does a lot of things out of principle and out of an intrinsic motivation of doing the right thing. This is something that I think is a very strong indicator on if you should choose something. Because, over time interest can dissolve. There are macro changes, and those within that can cause this. The lack of global perspective is too onerous. In a traditional setting, the needs/desires of the many are often overlooked. So, this is something that has to come from the outside.
The other thing is that I wanted to gradually build up the stones, the different cornerstones of my project myself. So, I basically built StakeNow because I wanted to engage into the regulatory discussions with our government, the financial authorities, and regulators.
Because, how can I use a blockchain if all the regulations underneath are not clear and nobody knows how the taxation is done? So, I take step by step all the different things that I think are relevant for regulation and I’m beta-testing them on a very simple use case — staking. The staking serves a very important use case, it’s maintaining the infrastructure and making infrastructure accessible and usable in Germany — which is the basis for a big company use case.
Were transactions per second (TPS) a factor within your decision basis?
No, completely irrelevant. I fully agree with Arthur, he wrote an article called Scaling Tezos. I think he’s right about a lot of things and you shouldn’t go with the hype. You should always go with the hard things first. For Tezos, it’s safety, language design, and architecture — all those things that are not seen at the beginning but, they pay off long-term.
As I have a long-term project, I intend to accompany the development process of the underlying technology with my growing project. I don’t need something that is here right now and functional one-hundred percent, I need something that can be functional one-hundred percent in the time that I need it. Also, I need a feedback loop to someone that actually cares and wants to make it happen.
Blockchain has a clear use case within proving the validation of autonomous driving functions. With the continued development of the ENVITED ecosystem work group’s prototype and continued work towards homologation — it makes for a key project to follow within the Tezos ecosystem.