Without AI you can’t be safe on the road

Petr Bakulov
Share

Petr Bakulov is a high-tech systems scientist, entrepreneur. He was Associate at Department of Transport Telematics and Lecturer in the Basics of Vehicle Management Automation. His research spans from automotive engineering (with a focus on vehicle management automation) to detection of car malfunctions by the sound the car is making. He is most known for his work in the field of Acoustic Fault Trace as a Diagnostic Parameter of Modern Vehicles at Scopus. Bakulov is a founder of V2M

Petr Bakulov speaks with Zia Askari from TelecomDrive.com about the key priorities for V2M today and how it is driving innovation in the market.

What are the key priorities for V2M today with wide spread of AI? 

That is a really good question. We are, in fact, the creators of the methodology. At present, the methodology is successful in detecting malfunctions in a vehicle. However, in the future, we want to develop our artificial intelligence, our neural network, to the point where we can predict a malfunction.

It is great that we now have the capability to understand if something has gone out of order. However, it’s much more feature-rich and thrilling to think that we can implement a system that says, “In 1,000 miles your wheel hub will have failed.”

And we are not only interested in working within the automotive market; we want to scale the solution to all possible fields that can use it, including manufacturing, robotics in factories, oil pipelines, gas leaks, and so forth. We are in contact with companies in different parts of the world, and we can confirm that it is not just the automotive industry that is interested in our technology.

Petr Bakulov

How do you look at the impact of AI in the automotive industry today?  

Artificial intelligence in the automotive industry today is primarily focused on autonomous driving. The release of the first serial self-driving autonomous vehicle is certainly an achievement that will transform the idea of transport, of mobility. In taxis and carsharing, we will see autonomous shuttles, which will be practically a two-in-one solution, because we will no longer need drivers. The regulations of the road will also change. It will really be a game-changing moment. Today, artificial intelligence, neural networks, and machine learning all focus entirely on autonomous driving. Moreover, an autonomous vehicle needs perfect on-board diagnostics. There is no way that such a vehicle can be safe, convenient, and user-friendly otherwise.

We hear about the banning of some AI based tools. So what do you think about this? Is there any risk of using AI in your market? 

In my opinion, today’s embargoes, the public letters, are primarily a competition story, because many of the ban’s signatories are developers of these technologies and are directly or indirectly engaged in the expansion and development of neural networks. At the same time, these solutions do make life easier and many processes better. In this respect, artificial intelligence is inevitable. In addition, it is absolutely unavoidable in the automotive industry, among other ways. As I said, autonomous cars will break up the world into a before and after and will ultimately change the concept of car ownership and use for all of humanity.

What are some of the innovations that V2M is driving in the market today? 

V2M is the perfect blend of automotive operating science and expertise in electrical engineering, electro-acoustics, neural networks, and machine learning. And just like in a well-made shake, all the ingredients, their combinations and execution, are essential in this case.

We are currently in the pre-seed stage, and most of what we do now is confidential information. What we can say is that we are working offline without using a server on board the vehicle. Neural networks require quite serious power from the hardware, and we have managed to make serious progress in inference mode. This is exactly the way a neural network operates, autonomously and without the involvement of a server or large processing power.

How are your solutions different from what other vendors are offering? 

We have no direct competitors today. As I previously said, V2M is a well-performed and intelligent mix with an excellent formula and a good bartender. Many solutions didn’t become widespread because they were vulnerable in terms of high cost or because the solution was unreliable in terms of low result validity.

Skoda had previously provided a kind of shazam-like service in the automotive field. However, the credibility of the solution was not up to par. Because modern drivers are not professionals, it is difficult for them to cope with such a service. Getting the sound right so that it can be properly processed later is quite a challenge, and it is a big risk for the end product.

The solutions that Kia and BMW developed in the 2000s were very expensive. Volvo and Mercedes had similar solutions, but they were more about vibration than sound and they were used exclusively as service solutions.

Our goal is to protect the manufacturer from false complaints by dishonest service providers and unscrupulous customers, even from those without intent; someone feels something and the specifics of the work may be mistaken for a fault and the buying experience spoiled. We defend dealers because a diagnosis that leads to no result (i.e., on a false symptom) is a direct loss, and of course, you want to get rid of that pain. Therefore, we work for all market participants—we help the manufacturer, the dealer, and the end consumer.

Please share your go-to-market strategy and your key target geographies?  

So far, we have only worked in the US, but we plan to go beyond that in the future, precisely because working with a car manufacturer is a worldwide effort. Our strategy for entering the market is to deliver to the factory and add our solution to the existing onboard diagnostics. This will make onboard diagnostics truly advanced, and with V2M, it will be 100% onboard diagnostics that can be relied upon and protects the car’s owner with confidence.

What are your plans for achieving growth in the coming months? Where is this growth going to come from?

We are ready for our first investment round. This is an important step for any startup, so we are no exception. This year we are planning to launch an MVP and our first pilot project. We are in negotiations with major partners and are in touch with various major car manufacturers.


Share