As the baby boomer generation ages we find our elder care system completely overwhelmed by the sheer amount of the people it needs to accommodate. A solution that a lot of tech people have come up with is elder care robots, but a lot of people are concerned with the emotional consequences of the elder being alone with the robot all the time, whole they be sad from the lack of interaction? How can we help that? We thought of Expressibot as a robot that would sympathize with the user and take the first steps to meaningful emotional interaction with robots. Expressibot is, what we believe, a strong start to human robot interaction in a way that most robots do not accommodate for yet.

What it does

ExpressIBot takes a regular Android smartphone and turns it into a lean mean feeling machine.

How we built it

We attached a phone to a 2 degree of freedom servo series, attached those servos to an arduino, added a raspberry pi to facilitate communication between and the machine learning processing system, used the same pi to connect to a pi camera that could take and forward the picture to the machine learning processing system, have said system read tweets to our dedicated twitter and take a picture to observe the user's reaction, and have the system send a command to the arduino to change the expression on the phone.

Challenges we ran into

The servo system we were

Accomplishments that we're proud of

Generation and implementation of a unique synonym mapping system for sentiment analysis, introduction to arduinos, using socket networking with java and python, creating an android app

What we learned

We learned that hardware is generally the limiting factor in a hackathon such as this. The servos and their housing were falling apart, causing us to abandon the movement aspect of the bot.

What's next for ExpressIBot

The future of ExpressIBot sees a proper chassis and full interaction mechanisms, and utilization in elderly homes and homes of the mentally challenged so that they may be able to lead a more fulfilling life with their social interactions.

Share this project: