AI-Based Self-Diagnostic Solution for Vehicles Makes One Hear Malfunctions to Improve Safety

-

Sponsored Post More Infos

V2M makes cars “hear” the sound of a malfunction and determine what exactly the problem is. The main idea behind this project is to increase safety on the roads. The company is already developing a scoring model that will be able to predict potential malfunctions by the sound of the car. V2M, a Delaware-based company, is the first in the world to develop a sound-diagnosis solution for vehicles and special equipment malfunctions. These methods make vehicles hear the sound of a malfunction and determine what exactly the problem is. So, vehicles will be able to diagnose a lot of specific malfunctions the moment they occur (or even just before it).

The idea was born back in 2012 when engineer Peter Bakulov decided to partially replace the work of a consultant advisor at the dealership with an automated expert diagnostic system. He wanted to implement this in his dealership business, but when he realized the potential of this idea, he made it a thesis of his Ph.D. work and began to form a science base for the product. It took nine years to walk a path from an idea to an utterly scientific case and a patent right in the United States. In 9 years of V2M history, almost a dozen articles were published, and one of them, “Acoustic Fault Trace as a Diagnostic Parameter of Modern Vehicles,” was published in Scopus. This is the highest level of scholarly publication, and it marks V2M recognition at the international scientific level.

V2M car
Image: V2M

The company has already installed the prototype in the world’s most sought-after electric car, the Tesla Model 3 Standard Range Plus. Tests are currently in progress. Their next step is purchasing two vehicles with internal combustion engines and one hybrid vehicle for product testing to show that the product is equally suitable for every type of car/engine.

How a vehicle’s acoustic self-diagnostics can save a driver’s life?

Insurance companies recently started the narrative about noises and the importance of noticing them immediately. Geico (listed in the top 5 largest insurance companies in the US) has published the article “7 Car Noises You Should Never Ignore“. The article lists all the noises that can signal trouble and gives a recommendation for drivers. Mike Peth, director of technical training at Ohio Technical College in Cleveland, recommends rolling down the windows and listening for any odd sounds and doing this every once in a while. He says the drivers know their cars and can pick up something that may become a problem. So, right now, drivers have two options: roll down the windows and listen or trust automated systems like V2M.

The data shows that 27% of malfunctions that led to accidents could be prevented if the noise car was making were noticed and recognized on time. In addition to that it is worth noting that monitoring and checking in case an emergency occurs might save a considerable number of insurance companies’ funds, not in the US only, but all over the globe. “In the past, I was a racing driver. In the summer of 2019 in front of my eyes, the car crashed and burned (I was two cars behind). Unfortunately, the driver died. His car had a  malfunction with the rear axle. It’s a common case and for sure it made a noise before finally jamming. Our solution could really save his life.”, said Peter Bakulov, CEO of V2M.

The ultimate goal for V2M is to work with automakers Tier-1 to create a one-time program implementation into the vehicle as a part of the standard equipment. Just like every car comes from a conveyor with a GPS system, the same way V2M plans to implement its system of malfunction recognition.

Tech details of the V2M solution

V2M system works using modern technologies in the field of electroacoustics and sound systems. V2M team has developed a methodology that, with the help of special software (artificial intelligence, multilayer neural network), determines the presence of patterns of faulty sound and concludes which malfunction is on board. The system recognizes a malfunction’s sound/noise even when the car is on the road loaded with noise.

For those interested in how this is possible.

Acoustic sensors consist of mems microphones with digital output pulse-density modulation (PDM). One sensor is located in front of the car, and the second is in the back. Both sensors are connected to the multiplexer board. We need a third sensor (in the middle of the car) if we want to understand what exact malfunction we have (this is their newest technology). The multiplexer board gives power to sensors and collects all data via Inter-IC Sound (I2S) protocol.

V2M solution
Image: V2M

System-on-a-chip

V2M device includes a system-on-a-chip (SoC) with a set of interface inputs (Inter-Integrated Circuit (I2C, I2S, CAN, Ethernet). All of that provides power, collects information from sensors, and combines data. It’s important to mention that our device has a microprocessor with a neural network coprocessor and an eSIM card (optional), so the car not only gets diagnosed almost in real time but also could collect and send needed information. In the MVP stage, some set-ups could be changed.

Algorithm

An algorithm was developed for this device, so it can collect and process data. It checks sensors periodically to ensure safe operation and diagnose any problems. The algorithm also enables us to add some new features that enhance the experience of users (e.g., automatic parking, and traffic jam assistance). With its help, we are able to detect failures in advance.

  1. Convert analog sound into digital and transfer digital data to a control unit.
  2. The software detects the sound and concludes what type of malfunction the vehicle has at the moment.
  3. The system notifies the driver (or somebody else, it depends on the client) about a malfunction and completes a report.

V2M also could be a 100% on-board system without any kind of M2M communications. Important, that we could not only detect malfunctions but also understand what exactly it is. In the end, 98% of the accuracy of recognition malfunctions, and it takes less than 2 seconds to recognize a malfunction. The company is already developing a scoring model that will be able to predict potential malfunctions by the sound of the car.

V2M can detect and report on vehicle malfunctions with 98% accuracy in just 2 seconds, and is being further developed to predict the likelihood of a malfunction before it occurs. This revolutionary technology is sure to revolutionize the automotive industry by helping drivers prevent their vehicles from breaking down. With V2M, drivers can be sure that they are always aware of what might be wrong with their vehicles, making it easier than ever to stay safe on the road.


Video: V2M promotional clip

Photo credit: The feature image is symbolic and has been taken by Andrii Vergeles. The images in the body of the article have been provided by V2M with permission to be used.

Was this post helpful?

Sponsored Article
Sponsored Article
This article has been sponsored and was submitted to us by a third party. We appreciate all external contributions but the opinions expressed by the author do not necessarily reflect the views of TechAcute.
- Advertisment -
- Advertisment -
- Advertisment -
- Advertisment -
- Advertisment -
- Advertisment -