There’s more to automation than a self-driving car. In the future, you may be able to avoid trips to a car garage altogether.

How? Future vehicles may be able to diagnose operational problems without human interference. What does that mean for you? Fewer trips to the garage for automotive diagnostics, for starters. Let’s dive into the details of these latest automotive technologies and see how your driving experience may change in the future.

You may be familiar with the tell-tale sound that something is up with your engine. In fact, sound plays a huge role in the diagnosis of engine faults. Both Hyundai and ŠKODA have developed AI-apps which process engine sounds to diagnose engine faults.

Experienced mechanics are often capable of identifying specific faults in a car, just from the sounds of an idling engine. They are attuned to the baseline sounds of well-maintained machines, and abnormal sounds are identifiable by ear. The practice of identifying faults by sound or vibrations is common using an automobile sound scope. Based on that principle of identifying specific sounds, Hyundai have developed an AI-powered fault detection and diagnosis system. 

Engine NVH Research Lab on AI

After years of research, the engineers at the Engine NVH Research Lab successfully extracted high quality data, based on technology that accurately establishes baseline and faulty engine sounds by category. In order to extract raw data that allows AI learning, the research engineers collected sound data from various parts of the engines that were fully functional, to fault-induced engines. When these collected sounds were processed, analysed, and categorized by tools, they became part of a growing database that the AI could learn from.

The engineers at the Engine NVH Research Lab have collected 830 sound samples which are further categorized under 18 types and 44 sub-types, based on the component and the nature of the fault. For example, a common fault type piston noise can be further categorized into piston ring noise, or piston friction noise, and so on. Once the AI learns enough variables to sound that an engine can make, it begins to recognize similar patterns of sounds, providing inferred diagnostics based on what its sound sensors “hear”. When AI diagnosis was compared to diagnosis by an experienced mechanic, only 8.6% of the experts made the correct diagnosis. AI accuracy was an overwhelming 87.6%. That is a solid ten-fold accuracy over the experts. AI accuracy will undoubtedly increase as more data is gained and better inferences can be made from it. 

ŠKODA IT and AI

Klaus Blüm, Head of ŠKODA IT, added, “At ŠKODA, we are consistently looking to comprehensively digitalise processes, products and services for our customers along the entire value chain. In order to be able to recognise the potential of innovations early on, develop them quickly and use them intelligently, we are continuously coordinating with the specialist departments to jointly implement new digitisation projects.”

The “Sound Analyser” app uses artificial intelligence to assess the current condition of wearing parts reliably, clearly, and quickly and notify technicians of any required servicing. For this purpose, the program considers various vehicle-specific parameters and analyses the usage profile of the respective car. To this end, Sound Analyser makes it even easier for technicians to perform accurate diagnostics on a vehicle as they will only need a standard smartphone or tablet to use the app.

Their smartphone app has been trialled in 14 countries – including Germany, Russia, Austria, and France – since June 2019. A total of 245 ŠKODA dealers have been taking part in the pilot project. The gradual introduction of technologies for determining any acoustic deviations from the norm will open up a wealth of new possibilities in terms of sensor-based, predictive maintenance in future. In addition, the vehicle’s online connection can be used to arrange an appointment directly with the responsible garage if necessary.

AI and Audio 

Audio interpreting AI apps to diagnose engine faults are just the beginning, and it’s likely that AI will play a huge role in fault diagnosis in increasingly complex vehicles. Luz Mauch, executive vice president of automotive at DXC Luxoft, explains.

“With increasingly complex software being brought into the vehicle, mechanics will not be able to just view an error code and look up a solution,” he says. “They will need support, either from an expert or perhaps from artificial intelligence (AI).” 

To deliver this support, vehicles will need to be enabled with the capability to be remotely diagnosed. The vehicle would be able to monitor its own status and request repair when it identifies an issue. Furthermore, analysis of the data that a vehicle or fleet of vehicles produces could allow faults to be detected before they impact performance. 

“As automakers place priority on customer experience, remote diagnostics will enable them to deal with issues in the vehicle with minimal interaction from the consumer,” says Mauch. “Having remote diagnostic services enabled therefore significantly improves the quality of service.” 

From engine faults, to detecting faults in the complex software running in modern vehicles, AI diagnostic solutions for vehicles are here to stay. In the future, if the car you purchase comes equipped with a self-diagnosing AI, there’s a good chance you’ll be able to check on the status of your tyres, engine, headlights, oil, and more. This means that you’ll be able to tell, early on, when it looks like something may be going wrong. If you can fix a problem with your car before it expands, you’ll be able to save on repair money – not to mention stress.

Finally

It’s also worth noting that self-diagnosing vehicles are likely to also be able to help drivers address environmental concerns they may have about their vehicles. AI apps which capture and process information on when your vehicle emissions are likely to be with us soon. In short, the future of self-diagnosing cars is broad and imaginative. While we’re yet to see widespread use of more modern self-diagnostic processes, it’s becoming clear that we’ll see more of them in the near future.