Keeping up on your car maintenance is crucial to make sure that your vehicle stays on the road.
In the past, maintenance relied on little more than the eye test, or friends who doubled as amateur mechanics. The industry transitioned to warning lights on the dash, and now our cars are riddled with costly sensors that require a professional technician to repair or replace.
MIT has created a new diagnostic software that could one day reshape preventative maintenance in vehicles. The software analyzes a car's sound and vibrations using a smartphone's microphone and accelerometers. It compiles the information and, using machine learning algorithms, displays a diagnostic rundown for the car in a smartphone app. According to the researchers, the app could save drivers up to $600 a year, and improve gas mileage — and that doesn't even include the potential savings from preventing breakdowns or towing fees.
For example, the app can tell if it's time to change the air filter by analyzing the sounds the engine makes. According to research scientist Joshua Siegel, the phone is "listening to the car's breathing, and listening for when it starts to snore." Actually, as the filter clogs, it starts to make a whistling noise that is indistinguishable to the human ear, but your phone can pick it up.
In order to monitor tire condition, the system ties into your phone's GPS and can tell if your tires are bald or overinflated by combining that GPS data with the car's speed and vibration readings.
The team is currently working on a prototype of the smartphone app, and they hope to begin field testing in the next six months. It should be commercially available in about 18 months.