Automotive diagnostic app analyzes a car’s sound, vibrations
The idea behind the app, say the researchers, is for users to be able to quickly get diagnostic information about their car, or any car they happen to be in, without knowing its history or having to connect to it directly in any way. Such diagnostic information could warn the driver of upcoming issues or required maintenance before they developed into a more serious condition potentially leading to a breakdown.
Using just a smartphone, the app is designed to be able to detect things such as if/when the tires need air, if the spark plugs are bad, or if the air filter needs replacing. According to the researchers, their smartphone app could save the average driver $125 a year and improve their overall gas mileage; for trucks, the savings could run to $600 a year, not counting the benefits of avoiding costly breakdowns.
The app was made possible, say the researchers, by the high sensitivity of today’s smartphones, which can detect relevant signals without needing any special connection. In some instances, however, mounting the smartphone on a car’s dashboard will further improve its diagnostic accuracy. So far, say the researchers, the accuracy of the results from the diagnostic systems they have developed are “well in excess of 90 percent.”
The researchers’ latest findings involved analyzing the need for replacing the car’s air filter. This is particularly tricky to diagnose, they say, since changing the filter too soon can be as bad as waiting too long. Knowing the optimal time can make a measurable difference in an engine’s performance and operating costs, and the app is able to detect this.
“We’re listening to the car’s ‘breathing,’ and listening for when it starts to ‘snore,'” says Joshua Siegel, one of the authors of a paper on the research. “As it starts to get clogged, it makes a whistling noise as air is drawn in. Listening to it, you can’t differentiate it from the other engine noise, but your phone can.”
The researchers tested the app using data from a variety of cars, some of which ran perfectly while others had deliberately induced issues ranging from a clogged filter to an engine misfire. Some of the diagnostics required a multistep process, using a combination of data collection and analysis.
For example, when checking for tire wear or overinflation, the system uses the phone’s GPS to monitor the car’s speed. Then, vibration data can be used to determine how fast the wheels are turning, which, in turn, can be used to determine the wheel’s diameter. That diameter can then be compared with that which would be expected if the tire were new and properly inflated.
Machine-learning processes were used to derive many of the diagnostics – comparing recordings of sound and vibration data from well-tuned cars with similar ones that had a specific problem. According to the researchers, some algorithms did a better job at detecting issues – such as wheel imbalances – than expert drivers from a major car company.
The researchers are developing a prototype smartphone app incorporating all the diagnostic tools and expect it to be ready for field testing in about six months. A commercial version should be available within about a year after that, say the researchers.
For more, see “Air filter particulate loading detection using smartphone audio and optimized ensemble classification.”
Related articles:
Automotive sensors market to grow at 6.7% CAGR to 2023
Turning cars into mobile devices
Connected car adapter company launches app store for cars
Volvo partners with Google to bring Android into cars
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
