During the last couple of days I have been working on a prototype to test a concept of my master thesis at Leiden University’s Media Technology program: Wearable devices that provide subtle autonomic feedback to their users. In this case we are talking about a glove that can measure its wearers heartbeat and vibrates along in the rhythm.
The pictures below provide some illustration of the current state of the prototype. It is a fairly simple and straight forward design that uses the awesome Pulse Sensor Amped, and a simple LilyPad Vibe Board, connected to an Arduino Pro Micro (deek robot) microcontroller boad.
The sensor is placed so that it can be fixed to the index finger of the left hand. Vibrations are generated by the vibe board at the wrist area over to the arteries located there.
The goal of this prototype was (1) testing technological feasibility, (2) evaluating the suitability of the involved design decisions for the next step of the research process, (3) and to provide me with some first hand experience of long-term autonomic feedback.
(1) has turned out to be very simple: Thanks to the great documentation and simplicity of the components involved, the device was functional in a small amount of time. Despite some difficulties while interfacing with the Arduino Micro, programming was straight forward. The Pulse Sensor folks provide well commented code examples for using their device with different Arduino versions, that lend themselves very well as a starting point for own projects like this one.
(2) However, it turns out that the sensor is not all that usable if one wants to gather heart rate continuously throughout the day. It uses pulse oximetry as a method for measurement, in which light of different wavelengths is passed through the skin. While the sensor is small and rather cheap, it has to be placed at thin areas of the body, such as the the fingers or the earlobe. Additionally, the user needs to hold still in order for the device to acquire accurate measurements. These two constraints make it less than perfect for an unobtrusive device, that should provide feedback for a longer period of time. The vibration module works like a charm and is neither too strong, so that it is bothersome, nor too weak, so that the feedback is not noticed at all.
(3) I have been wearing the device for several hours at a time this far. While the vibration feedback seemed strange at the beginning, I have adapted fairly quickly and stopped noticing it consciously after a while. Due to sensor-induced jumps in the pulse pattern when I move the hand, I have not been able to experience the device for a longer period of time. A modification of the device (or an entirely new prototype) could generate bogus pattern to simulate different conditions in autonomic feedback.
Update 14-09-2014: I experienced some additional issues while I have continued to test it. (1) the fact that it comes in the form of a glove already puts severe limits on the contexts in which it can be worn without being a nuisance to its wearer. Typing and other everyday activities are impacted a lot, because the sensor for measuring hear rate is located on the index finger. Additionally, the fiber of the glove — thin wool in this case — makes interacting with objects, especially food, difficult. (2) The used 9V block battery feels noticeably heavy after some time. (3) This prototype does not have an on/off switch integrated, but the device would greatly benefit from one. Perhaps a passive solution that turns the device on automatically once it is worn would be an interesting solution.