Adorable Peek-a-Boo Robot

For many years now there has been work on “affective” robots, which has produced some interesting and also some creepy results. Lacking natural language abilities, much of the work has explored facial expressions and temporal coordination (i.e., responsiveness).

This is not an area where I have much expertise, except as a consumer faced (literally) with such interfaces.

To date, I’ve been struck that many of the robots do not appeal to me. This was apparent to me in the case of “Jimmy” from twenty first century robot, which presents a child-like face that may appeal to kids but doesn’t to me. Jimmy is hardly alone, almost all “cute” robots just flat annoy me.

It seems clear to me that whatever “cute” means to specific robot designers, it isn’t the same for everyone. And, I’m sure, many of the devices are built by hackers who know and love the technology but have no particular expertise or interest in human social behavior. That’s fine, but its not the way to really build attractive robots.

Getting around to the point, I was impressed by a little robot, Mira, that is actually kind of attractive. Utterly useless, and displaying the intelligence of, say, a newborn human, It has a gloriously simple face, and elegantly simple movements.

Behaviorally, Mira does a couple of thing well. It tracks faces, particularly eyes, and orients its own “eyes” to peer at the person. (Just like a baby mammal.)

It is reported that Mira also is learning about “human emotion”, probably from facial expressions and analysis of vocal behaviors. (I’m not sure if Mira does speech recognition, but it could.)

But the key thing is that Mira likes to play a simple game of peek-a-boo, which both human and robot “understand” and “enjoy”.

Again, all of this is about the cognitive range of a newborn infant. But infants are highly attractive to human adults. This is a great thing to model, if you want a robot that people want to play with.

I think Mira is successful (for me) where others fail, may be due to the fact that the hacker in question, Alonso Martinez works at Pixar. This background explains the cinematic expressiveness, and the beautifully understated face and motion. And it also explains part of why you like Mira – it has a personality, just like cartoon characters.

So, Mira and Jimmy were developed with pretty similar design philosophies and absolutely the same technical capabilities. Yet I hated Jimmy and liked Mira. What is going on here?

I think the biggest factor is what kind of interaction happens. Jimmy is imagined to have amazing, near human conversations, as well as complex to the point of psychiatric emotional development. Mira can’t talk, but it plays peek-a-boo well.

This isn’t really any deep philosophical thing, it’s just a matter of doing a good job at something simple versus a weak job at something complex.

Plus, Mira delivers more than it would seem to promise, which Jimmy promises too much and disappoints.

Now that’s something show business people grok.


Robot Wednesday

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.