ARIA: Robots with Adaptive Software for Autism Therapy (And why I want to hack them!)

The Dude goes to a school that has an autism unit. Apparently that's a pretty rare feature among schools in our city. What makes it even more special is the way they get all the kids interacting on a regular basis. This year, the Dude's teacher organized a fundraiser: The Dude's class set up a company, voted on cookie dough flavours and then worked alongside the kids from the autism unit to make the dough themselves. It was such a huge success the first time, that they are in the midst of their second round of sales. Funds raised were used to buy items for a sensory room (think ball pit, lighting, bubble tube, etc). A phenomenal experience for all involved.

But this.

Robot using ARIA (video still)
What if there was a way to raise enough funds to bring in robots to engage students with Autism Spectrum Disorder, and assist in their therapies? Seriously! It would seem that children on the autism spectrum respond better to instructional cues from a robot than from an individual, or in this case, a therapist. This can be an important distinction in the early development of social skills as robots can
help children... learn how to coordinate their attention with other people and objects in their environment. This basic social skill is called joint attention. Typically developing children learn it naturally. Children with autism, however, have difficulty mastering it and that inability can compound into a variety of learning difficulties as they age. (Source: Science Daily)
A group of researchers at Vanderbilt University have now designed software to facilitate robot therapy, making it more adaptive and responsive. ARIA, the Adaptive Robot-Mediated Intervention Architecture, is designed to enable a robot to receive external input in real time from cameras located around the room and on a cap worn by the child. If the child is not responding to the instructions, the system can react accordingly: by tracking the child's head movement, if the child does not look where the robot tells them to look, it can provide additional or repeat instruction.

Robotic therapy isn't intended to replace therapists, but it can bridge the gap. And repetition doesn't get tiring to a robot; although the therapist might need a break :)

This is a fantastic use of technology. But it could go further. I think that this solution needs to be hacked to reach more people. Here's how:
  1. The researchers release their software code as open source so that it can be hacked and improved upon by the development community at large.
  2. Then, the dev community figures out how to hack off-the-shelf programmable toys like Robosapien so that the software can be uploaded and used in multiple environments (think cheap robot, some webcams and a couple of LED lights on a hat, and you've got a cheap system for home or schools).
And suddenly it's a solution that is available to the masses. Wouldn't that be amazing? A girl can dream, right? Meanwhile, check it out in action below:



(Source: Megagadget)


FYI: Join me this Friday, April 12th at the Ottawa Public Library Main Branch Auditorium for a FREE lunchtime session: "Hacking is a mindset, not a skillset." For more info and to RSVP, check out the event details.