How might we create a safer experience for cyclists using the appropriate amount of technology?

SAIVE (smart artificial interactive vehicle ecosystem) is a near-future vision where vehicles and commuters talk to one another while automatically recording and broadcasting data. I worked with Ananda Annisa Prasetyanto and Qiao Yin on this project. Special thanks to Carly Burton at Cognizant.

Concept Details

Turn Signals

If SAIVE predicts that a car will impact a nearby person, their turn signal will automatically flash even if the driver hasn't turned on their blinker. The key is that false positives have minimal consequence.



"Dooring" is when the opened door of a parked car collides with a biker. If SAIVE predicts this will happen, a notification will be projected onto the car window and the door will be temporarily locked. 

Technology behind SAIVE


Halo'd Smart Phones

A "halo'd" object is similar to a smart object, constantly producing data about it's environment. In response to passionate interviews, we intentionally chose to halo the smart phones of cyclists instead of the bikes themselves. Location and velocity data can be captured by a smart phone just in the pocket or backpack of a cyclist. 


Driver Data

Drivers already have the potential to produce a large degree of data through normal driving activity. Data can be collected on the speed of the car, angle of the steering wheel, weight on the seat cushion, and many more variables in order to accurately predict driver behavior.



DSRC, "dedicated short range communication", is similar to bluetooth technology and is the backbone of SAIVE. The implementation of DSRC within the ecosystem requires a vast degree of edge computing which isn't possible with today's technology. Edge computing is computing which occurs close to the source of the data itself, instead of being sent to a centralized location then sent back. 

Technology Already in Cities

Autonomous vehicles collect vast amounts of data, adding a wealth of information to SAIVE. Furthermore, smart traffic signals, successfully implemented by Stephen Smith at Carnegie Mellon, would contribute data to strengthen the ecosystem.


Our team conducted interviews with commuters (cyclists, motorists, and pedestrians) as well as experts in cycling and traffic flow in order to gain insights on the current transportation system. Mapping a cyclist's experience, ideation with the "crazy-8's" methodology, and gauging potential design solutions were part of the design sprint which led to the formulation of the AI ecosystem SAIVE.