Google is pitching its Activity Recognition Transition API as the key to unlocking insight about smartphone owners’ movements. The idea is to let developers decipher between simple movement, such as walking, and complex movement, such as driving all while maximizing battery life.
Google first debuted the Transition API in November on the Pixel 2 smartphones. It is what powers the phone’s Driving Do Not Disturb mode. It uses the phone’s sensor to gauge when phone is actually in a moving car, as opposed to engaged in some other movement.
“While it might seem simple to turn on Do-Not-Disturb when car motion is detected by the phone’s sensors, many tricky challenges arise in practice,” explained Google’s Marc Stogaitis, Tajinder Gadh, and Michael Cai in a blog post. “How can you tell if stillness means the user parked their car and ended a drive or simply stopped at a traffic light and will continue on? Should you trust a spike in a non-driving activity or is it a momentary classification error?”
The Transition API will give developers access to the same information Google has used itself to power DND. Google says developers can use its training data and algorithmic filtering to more accurately detect changes in user activity and reach the proper conclusion as to what the person is most likely doing. Perhaps most importantly, this means developers will not have to create their own solutions, which might not be as efficient as Google’s. Tapping into Google’s data should help reduce the power needed to make movement assessments and take action.
Right now, the API can help determine the difference between walking, running, cycling, driving, and sitting still. More features, such as determining the difference between rail vehicles and road vehicles, are on deck.
Google has been testing the API with QuickBooks and Life360 and both were impressed with the results.
“Before the Transition API, we created our own solution to track mileage that combined GPS, phone sensors, and other metadata, but due to the wide variability in Android devices, our algorithm wasn’t 100% accurate and some users reported missing or incomplete trips,” explained Pranay Airan and Mithun Mahadevan from Intuit. “We were able to build a proof-of-concept using the Transition API in a matter of days and it has now replaced our existing solution, offering a more reliable solution that also reduced our battery consumption.”
Google says it will continue to add features to the Transitions API over the coming months in order to support more context-aware functionality. Details concerning how to put the Transitions API are available here.