Mobile devices such as tablets and smart watches could become quicker to charge and slower to run out of juice thanks to a new approach to designing batteries from Microsoft researchers.
The battery inside any given gadget today typically has one particular chemical design that defines its capacity and how quickly and efficiently it can charge and discharge. Our devices might be much improved if they instead had multiple smaller batteries, each with slightly different chemical makeups and performance, says Ranveer Chandra, a principal researcher at Microsoft.
Software could then choose how much to charge and discharge a device’s different batteries at any moment based on what the device was being asked to do—or what it might expect to do in the future, based on a person’s typical activity. “This can allow the battery to become a more intelligent, customizable entity,” says Chandra.
For example, a phone might have one battery that can efficiently provide a lot of power that handles an activity such as playing games during your morning subway ride, and a second battery with a lot of capacity that’s good at trickling out power over long periods of idling time.
Chandra says that “software-defined batteries,” as the researchers call them, could also help drones and electric cars. A car might be able to use cues like the destination punched in to its navigation system to figure out how to manage batteries of different properties, says Chandra. The route, inclines, and expected traffic conditions could all help the car use its energy more effectively, he says.
The Microsoft researchers have tested their idea by modifying real mobile devices such as phones and tablets. They modified the software on the devices, and linked them to a custom circuit board that could manage four different external batteries.
One test was intended to determine how a smart watch could benefit from having a flexible battery in its strap as well as a conventional battery in its body. Bendable batteries are not very efficient at providing a lot of power at once. So the researchers programmed their software to carefully balance the work of the two batteries during a simulated day of activity that included checking messages and going for a run. In the test, that balancing act extended battery life by more than an hour.
Results of those experiments will be presented in a paper at the Association for Computing Machinery Symposium on Operating Systems Principles in Monterey, California, this week.
The concept of software-defined batteries emerged from a research effort at Microsoft that began in 2012 and is aimed at dramatically increasing the battery life of mobile devices (see “Microsoft Aims for Phones That Last a Week”). Although software-defined batteries are just a research project for now, Chandra has worked with colleagues from Microsoft product groups working on Hololens and the Surface tablet.
With battery life a major shortcoming of wearable devices, the software-defined battery approach could gain traction in the industry, says Christine Ho, cofounder and CEO at Imprint Energy, a company developing flexible batteries (see “Flexible, Printed Batteries for Wearable Devices”). And being able to combine new battery technologies like Imprint’s with older, better understood ones could make it easier for the newcomers to become commercially successful and change what devices can do.
However, putting multiple batteries and extra components and software into devices doesn’t come for free, she notes. “For high-end devices it makes a lot of sense, but it may not be possible for all types,” says Ho. “For many wearables they will need to be low-cost, and space is very constrained.”
Going bald? Lab-grown hair cells could be on the way
These biotech companies are reprogramming cells to treat baldness, but it’s still early days.
Meet Altos Labs, Silicon Valley’s latest wild bet on living forever
Funders of a deep-pocketed new "rejuvenation" startup are said to include Jeff Bezos and Yuri Milner.
Meta’s new learning algorithm can teach AI to multi-task
The single technique for teaching neural networks multiple skills is a step towards general-purpose AI.
A horrifying new AI app swaps women into porn videos with a click
Deepfake researchers have long feared the day this would arrive.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.