You’ve had a couple of cold ones, then head to your car. Suddenly, your iPhone starts shrieking out an alert. It’s telling you that you’re too impaired to get behind the wheel.
This is the goal of AlcoGait, which uses a smarthone’s built-in sensors to monitor the way the user is walking. AI is used to detect if the person’s gait indicates they’re not up to driving.
The software is in development by Worcester Polytechnic Institute (WPI) and Boston University. Costs are currently being paid for by $500,000 grant from the National Institute of Health’s National Institute on Alcohol Abuse & Alcoholism.
The head of development is Emmanuel Agu, an associate professor of computer science at WPI. “When people drink and they’re over the alcohol limit, basically one of the first parts of the brain to go is the one that tells you you’re too drunk to function,” Agu told Metro.us. “So, many people are too drunk to drive, but they don’t know that they’re too drunk.”
AlcoGait constantly looks out for you
There are apps available now that ask you questions and give you hand-eye coordination tests to try to tell if you’re too impaired to drive. Druid is one example. These are digital equivalents of police asking you to stand on one foot while reciting the alphabet backward.
The disadvantage of these is the user has to remember to take one whenever they’ve been drinking. And they can potentially be fooled by moderately intoxicated people concentrating long enough to pass a short test. That doesn’t indicate they’re really up driving.
The goal of AlcoGait is to be constantly running the background. It will learn how you walk when you’re sober, giving it a baseline for comparison when you’ve been drinking.
Boston University professor Michael Stein explained to The Daily Free Press, “If the phone app can detect drunkenness when you walk, it also can alert you that you shouldn’t be driving and that you’re at risk for falling, injuring yourself, or doing other risky behaviors,
Eventually, the software will be able to automatically deactivate the user’s car and hail a ride-sharing service.
Not available yet
Clinical testing began in October of 2017. This involved getting volunteer college students drunk, checking their blood alcohol level with a breathalyzer, then seeing if AlcoGait could determine if the test subject was intoxicated.
Agu’s team originally hoped to have AlcoGait out by the end of last year. Perhaps some tweaking was required after testing. Hopefully, the release will come soon.
