Be Seen; Be Safe.
Jake Scherlis, Jonathan Ortiz, Myles Blodnick, & Kaan Doğrusöz
Before the advent of turn signals, automobile drivers would use their left hand to signal turns. Cyclists adopted this, using these same hand signals to turn and change lanes. However, as electric turn signals have become commonplace, many new drivers are unfamiliar with the use of hand signals, leading to accidents and dangerous driving patterns.
The goal of our project was to enhance the communication and clarity of a bicyclist's hand signals without hindering the riding experience or requiring any extra input on behalf of the bicyclist. This greatly improves the bicyclists' safety by making their turn signals more easily interpreted by automobile drivers. With the use of this glove, the association between hand signals and vehicular turn signals is restored.
For our first iteration we designed an analog gravity sensor that enabled the activation of LED strips without any digital computation.
The second working prototype is a massive improvement over the original. We improved our design by mounting the system onto a flexible fabric base that allows it to fit over existing bicycle gloves. Cyclists would not have to give up their existing bicycle gloves and can instead use the signal as an accessory. The fabric itself can be worn as a glove in its prototype form, but in a future iteration we’d like to explore a more minimal, permanent casing.
To improve visibility, we elected to use much brighter NeoPixel strips that contain individual drivers for each LED, which we utilized to add animation to our signal. The animated arrows provide more immediately understandable information to people looking at the cyclist by providing context for what the lights on the cyclist’s left hand symbolize.
We improved the performance over our first iteration by replacing our rudimentary analog sensor with an accelerometer. When we first started testing, we figured that x, y and z axis measurements would be sufficient to determine the position of the glove and therefore which sensor to activate, but during actual trials we found that accelerating and decelerating significantly affected the ranges in which the glove would select output signals. We had to adjust ranges to compensate. The current iteration of the glove is a huge leap over our analog tilt switch iteration as a result; it is unaffected by elements such as wind and is far less prone to spontaneous failure.