Cars are becoming better and smarter at interacting with the world outside, but not so
much with the people inside. We see an opportunity to improve the experience of the driver
and all the occupants of the vehicle, in the same way products such as Amazons echo have done
for the home.
We propose an intelligent assistant that can respond to and predict the users needs. Our system is
aware of exactly who is in the car and what the sentiment is at any given time. Using such data,
we can learn behavior about the driver and ultimately improve the driving experience by predicting
the drivers needs while at the same time reducing distractions.
Use case: a mother, her child and their dog get into the car on a Friday afternoon. Detecting the
presence of all the occupants, our system has learnt that there is an extremely high probability their
destination is the park. The route with the least traffic is now suggested and a video is put on in
the back seat for the child. A spotify playlist is also suggested to the mother.
We offer our software directly to manufactures to integrate into their vehicles, or can work with
manufactures to develop an app that can be integrated to work with the software of the vehicle.
Using machine learning algorithms such as neural networks allows us to monitor video in realtime to
detect the presence of anything we choose to train the system on. The visual and sentiment
features along with the driving data are then again used to predict driver behavior. Every time an
occupant accepts or declines the suggestion, our system learns. It's this behavior of the
system that allows us to offer an experience fully tailored to whoever is in the vehicle.