MIT team develops software that can tell if tires need air, spark plugs are bad, or air filter needs replacing.
Imagine hopping into a ride-share car, glancing at your smartphone, and telling the driver that the car’s left front tire needs air, its air filter should be replaced next week, and its engine needs two new spark plugs.
Within the next year or two, people may be able to get that kind of diagnostic information in just a few minutes, in their own cars or any car they happen to be in. They wouldn’t need to know anything about the car’s history or to connect to it in any way; the information would be derived from analyzing the car’s sounds and vibrations, as measured by the phone’s microphone and accelerometers.
The MIT research behind this idea has been reported in a series of papers, most recently in the November issue of the journal Engineering Applications of Artificial Intelligence. The new paper’s co-authors include research scientist Joshua Siegel PhD ’16; Sanjay Sarma, the Fred Fort Flowers and Daniel Fort Flowers Professor of Mechanical Engineering and vice president of open learning at MIT; and two others.
A smartphone app combining the various diagnostic systems the team developed could save the average driver $125 a year and improve their overall gas mileage by a few percentage points, Siegel says. For trucks, the savings could run to $600 a year, not counting the benefits of avoiding breakdowns that could result in lost income.
With today’s smartphones, Siegel explains, “the sensitivity is so high, you can do a good job [of detecting the relevant signals] without needing any special connection.” For some diagnostics, though, mounting the phone to a dashboard holder would improve the level of accuracy. Already, the accuracy of the results from the diagnostic systems they have developed, he says, are “all well in excess of 90 percent.” And tests for misfire detection have produced no false positives where a problem was incorrectly identified.
The basic idea is to provide diagnostic information that can warn the driver of upcoming issues or needed routine maintenance, before these conditions lead to breakdowns or blowouts.
Take the air filter, for example — the topic of the team’s latest findings. An engine’s sounds can reveal telltale signs of how clogged the air filter is and when to change it. And unlike many routine maintenance tasks, it’s just as bad to change air filters too soon as to wait too long, Siegel says.
That’s because brand-new air filters let more particles pass through, until they eventually build up enough of a coating of particles that the pore sizes get smaller and reach an optimal level of filtration. “As they age, they filter better,” he says. Then, as the buildup continues, eventually the pores get so small that they restrict the airflow to the engine, reducing its performance. Knowing just the right time to replace the filter can make a measurable difference in an engine’s performance and operating costs.
How can the phone tell the filter is getting clogged? “We’re listening to the car’s breathing, and listening for when it starts to snore,” Siegel says. “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.”
To develop and test the various diagnostic systems, which also include detecting engine misfires that signal a bad spark plug or the need for a tune up, Siegel and his colleagues tested data from a variety of cars, including some that ran perfectly and others in which one of these issues, from a clogged filter to a misfire, was deliberately induced. Often, in order to test different models, the researchers rented cars, created a condition they wanted to be able to diagnose, and then restored the car to normal.
“For our data, we’ve induced failures [after renting] a perfectly good vehicle” and then fixed it and “returned the car better than when we took it out. I’ve rented cars and given them new air filters, balanced their tires, and done an oil change” before taking them back, he recalls.
Some of the diagnostics require a complicated multistep process. For example, to tell if a car’s tires are getting bald and will need to be replaced soon, or that they are overinflated and might risk a blowout, the researchers use a combination of data collection and analysis. First, the system uses the phone’s built-in GPS system to monitor the car’s actual speed. Then, vibration data can be used to determine how fast the wheels are turning. That in turn can used to derive the wheel’s diameter, which can be compared with the diameter that would be expected if the tire were new and properly inflated.
Many of the diagnostics are derived by using machine-learning processes to compare many recordings of sound and vibration from well-tuned cars with similar ones that have a specific problem. The machine learning systems can then extract even very subtle differences. For example, algorithms designed to detect wheel balance problems did a better job at detecting imbalances than expert drivers from a major car company, Siegel says.
A prototype smartphone app that incorporates all these diagnostic tools is being developed and should be ready for field testing in about six months, Siegel says, and a commercial version should be available within about a year after that. The system will be commercialized by a startup company Siegel founded called Data Driven.
Learn more: Let your car tell you what it needs
The Latest on: Machine learning systems
- How E-learning Platforms are Becoming a Disruptor for Successful New-age Hiringon August 17, 2019 at 12:56 am
machine learning, AI and data engineering. With the current education system completely out of sync with the changing requirements in the professional world, E-learning platforms offering programs on ...
- Learn AI and deep learning with this bundle, for a price you chooseon August 16, 2019 at 11:15 am
You’ll learn how to implement deep learning algorithms for machine learning systems and integrate them into your product offerings, including search, image recognition, and language processing. This ...
- Y Combinator-backed Holy Grail is using machine learning to build better batterieson August 16, 2019 at 9:08 am
“There are some companies that only do the machine-learning part and the computational ... Holy Grail pitches a system that can get real test batteries into the hands of end customers in the ...
- Machine learning brings cell imaging promises into focuson August 16, 2019 at 2:54 am
So we need to pressure test the system,” says Boyd. There are also broader hurdles to overcome. For one, drug hunters may need to embrace a different way of thinking to appreciate the patterns that ...
- Machine learning reveals links between genetic factors and behavioron August 15, 2019 at 8:40 am
While mice forage for food, various different neural systems are involved in controlling seeking behaviors, navigation, memory, anxiety, hunger, attention, and reward. Using the machine learning ...
- New Partnership Uses Machine Learning to Improve Data Securityon August 15, 2019 at 6:45 am
This will enable the design of machine learning tools that can analyze large datasets, determine what additional data could be collected to potentially improve analysis, and uncover possible security ...
- The Danger of Over-Valuing Machine Learningon August 14, 2019 at 2:48 pm
Getty There are four issues that arise with basic machine learning. The first is that you need to go through the process initially of training the algorithms by identifying a large enough (and ...
- Five Breakthroughs In Machine Learning Marketers Should Know Abouton August 14, 2019 at 2:33 pm
Their system combs through tweets to find mentions of vulnerabilities, and then uses a machine-learning algorithm and natural language processing (NLP) to determine how much of a threat each one is.
- DARPA Selects BAE Systems to Develop Machine Learning Capabilities for Space Situational Awarenesson August 13, 2019 at 10:08 am
BAE Systems has been awarded a Phase 2 contract to develop machine learning capabilities aimed to help the military gain better awareness of space scenarios for the U.S. Defense Advanced Research ...
- BAE wins DARPA contract to develop machine learning technology for space operationson August 13, 2019 at 6:42 am
The second phase focuses on the use of machine learning for space and situational awareness, the company announced Aug. 13. BAE Systems originally won the contract for the Hallmark Tools ...
via Google News and Bing News