The Open Source Ring-leader

Smart rings have gained significant popularity since the miniaturization of electronic components that made them possible. These technologically advanced rings combine fashion with functionality, offering users a wide range of capabilities right at their fingertips — literally. Smart rings can be used for various purposes, including fitness tracking, mobile notifications, contactless payments, and even controlling smart home devices. Despite their small size, these devices are often equipped with sensors and wireless connectivity, allowing users to stay connected and monitor their health and lifestyle seamlessly.

If you want to hack away at a smart ring, you might find yourself with few options, however. Presently available rings are either commercial and proprietary, with the hardware and software design details undisclosed, or they are DIY projects with bulky form factors that are not practical for daily use. This limited accessibility to the inner workings of smart rings has posed a challenge for enthusiasts and developers seeking to explore and customize these devices to their fullest potential.

A new option has recently been developed by a team of engineers at The Pennsylvania State University that is both open source and compact. Called OmniRing, this smart ring platform has a miniature form factor, long battery life, sensing and processing units, and is water resistant, making it practical for daily use. It was designed with finger motion analytics and healthcare applications in mind, but the open design allows each user to customize it as much as they wish. And with the total cost coming in at less than $25 wholesale, or about $62 for a single unit, OmniRing is accessible to just about everyone.

The ring is composed of a flexible printed circuit board with an inertial measurement unit (IMU) and photoplethysmography (PPG) sensors. The Nordic Semiconductor nRF52832 system on a chip, with an Arm Cortex-M4 processor running at 64 MHz was selected to handle onboard processing. Since this chip also supports wireless protocols like Bluetooth Low Energy, it can transmit sensor readings to external devices for more complex analyses. The components are housed in a waterproof 3D printed case that was designed using a combination of resin and thermoplastic polyurethane. A 3.7 volt ring-shaped rechargeable LiPo battery, which can operate for up to a week between charges, completes the design of the 2.5 gram OmniRing.

The team envisioned a number of use cases that their smart ring would be well-suited for. Its sensing capabilities make it possible for the ring to monitor activities like running, eating, drinking, and sleep cycles. Viewing trends in this sort of data could provide the wearer with insights that help them to adopt an active and healthier lifestyle. Noting that OmniRing can also measure heart rate, blood pressure, and emotional states, the researchers believe that it could benefit people with certain medical conditions, like heart disease or high blood pressure.

One of the guiding principles behind the design of the OmniRing is that it be both usable and socially acceptable to wear. To assess the perceptions of wearers of the device, a small user study with 12 participants was conducted. The participants were given an OmniRing, as well as three other finger-based sensing devices, namely Tapstrap, Myo, and CyberGlove. They were then asked to go about their normal, daily activities while wearing each of the devices. At the conclusion of the study, OmniRing was given the highest scores in comfort, weight, and appearance.

The hardware was also assessed to determine how well the 3D finger tracking capabilities worked. The IMU data was leveraged for this purpose, and fed into a machine learning model that was trained to predict the location of the finger in 3D space. A joint position accuracy of 6.57 mm with a joint angle accuracy of 8.68 degrees was achieved using this approach. However, the algorithms did need to run on a smartphone, so the ring was not entirely self-contained in this case.

Looking to the future, the team hopes to add additional sensors to the ring, like electrocardiogram and electrodermal activity sensors. They are also planning to explore running their analysis algorithms on the device itself, instead of a smartphone, because the onboard Arm Cortex-M4 processor should be capable of handling these workloads. But they do not intend to do all of the work themselves — with the open design of the OmniRing, they hope to see extensions to the platform developed by the community.