IE 11 is not supported. For an optimal experience visit our site on another browser.

The human behind this year's hot robot

Robotic physicist Mark Tilden answers questions about his creation, Robosapien, and what's next in biomorphic robotics.
Robotic physicist Mark Tilden takes the controls of his creation.
Robotic physicist Mark Tilden takes the controls of his creation.Wow Wee

Robosapien took its curmudgeonly place this year in a toy menagerie that also includes cyber-critters ranging from the occasionally annoying talking Furby pet to the gleaming, expensive Sony Aibo robotic dog.

The humanoid robot stands out in part because there's a human behind the marketing campaign: Mark Tilden, the British-born "robotics physicist" at Hong Kong-based Wow Wee. Tilden says he has worked for "NASA, DARPA and JPL through Los Alamos National Laboratory, and other government and private research agencies studying robotic methods."

Wow Wee's press material credits him with developing the basics for "biomorphic robotics" around 1988.

"I actually saw Mark Tilden on TV one night on the Discovery Channel in Hong Kong (about five years ago) and took the initiative to call him up and send him a ticket to come to Hong Kong to meet with me," the company's president, Peter Yanofsky, told an interviewer. "The rest is history — he never left. Now he is like family. It's very rare and lucky in life to be able to meet people like Mark"

Before Robosapien, Tilden created the B.I.O. Bug toys for Wow Wee, and he also served as a consultant for robot scenes in the movie "Tomb Raider." After Robosapien, his fans hang on his every word hinting about the next generation in biorobotics. Last week, during his travels in Europe, he answered a few e-mail questions from MSNBC.com:

Q: Can you give a few more details on the past work you've done in biorobotics? You have referred to work with robotic bugs — does that relate to the Entomopter or any other project that's still extant?

Tilden: My science work concentrated on the evolution of minimal real-world robots from the lesser to the greater, under the assumption that robots should evolve by themselves for themselves. Started out with solar amoeba, latest is the Robosapien.

Most of my science stuff was highly controversial because though it worked, it was not built in conventional fashion so lacked the predictablity to be politically profitable. Few colleagues studied or used it. Still, that was no problem for the toy industry. Though I still write the occasional science paper, the proctored evolution of biomorphic robots is now in a different survival environment, namely the common market.

Robosapien is not a devolved human, he's an evolved robot, that burps, for money.

Were there any lessons learned from the Furby experience?

Tilden: That wasn't our experience, but the industry-wide message is that people will buy robot things if they appear "alive" enough.

Are there any Robosapien hacks that you've found to be particularly clever?

Tilden: So far, only standard hacks have been implemented, like cameras and PDA brains. There are several other hacks I could mention, like 10-farad supercap power regulation to stabilize his strength, or that his hands are removable, exchangeable and modifiable, or that the LEDs in his hands can be replaced with infrared ultrabrights so he can use them as visual flashlights, or that neodenium boron supermagnets under his shell would allow for mountable tools, armor or weapons (most 1/60th-scale Gundam accesories fit him fine), or that you could swap out his motors with high-power 130 types and separate H-bridge drivers to triple his speed and strength, or that his chest has more than enough room for a laser-guided pan-and-tilt 2.21-GHz color camera targeting system that you can patch into his reconfigurable reflex sensors.

But doubtless these will come in time. Until then, the best hacks I've seen are the ones where people had real fun with their robot, like using him to throw bread at pigeons, or dressing him up for rock videos, or kicking a can around a busker street dancing occasionally for battery money.

That's the nice thing about building devices like this: Toys are good news. Applications are not just for science.

What's your strategy on product cycle?

Tilden: Reminding people that Santa Claus don't wait.

It does sound like you and your colleagues are acutely aware that Robosapien will have a limited cycle and that the "next big thing" is already in the works. Any more details you can share on the next generation?

Tilden: We're expanding and maturing our robots' family line. We're hoping to do for vision what we've done for walking. That's all I can say.

Do you feel as if your dream designs had to be "dumbed down" or "cuted up" to fit technical or marketing realities?

Tilden: In this case no. We had to make compromises, but this item is pretty much what I wanted — universally acceptable but secretly hackable. Looks cool, too, without being too juvenile.

I read somewhere that the design is innovative in that it's based on triangles rather than squares. If that's so, could you explain the geometry and the balance that's involved?

Tilden: Yes, but it's complicated. Let's just say more roboticists should read Buckminister Fuller.

What other innovations are you particularly proud of?

Tilden: Power efficiency, reliability, fluidity, hackability, and all with toy-grade components and pricing. A combination of effects that made for an inexpensive but widespread multipurpose mechanical mobile minion.

Are there things that you believe your former employers (NASA, DARPA, Los Alamos, etc.) could learn from this?

Tilden: I'm hoping a few will try out their many Artificial Intelligence and Artificial Life science ideas using the Robosapien. Many colleagues complained there was always a shortage of inexpensive robot bodies to do experimentation with. The Robosapien is smart enough to use in an advanced robotics course as is, but with a little tweaking...

You've talked about going after older users with future products. What sorts of features help you target that market?

Tilden: Targeting grownups is a high performance-to-action ratio; press one button to get a lot. But the best design guide is to make something that entertains without patronizing the user, and with some secrets tossed in to keep it interesting.

Do you see Robosapien as remaining part of a line that also includes these future products, or do you expect that RoboSapien will be phased out of production and overtaken by more advanced, more "mature" models?

Tilden: That depends on the market, of course. We are planning on newer models having control over their older brethren, but each will have their own distinct charms and advances over this first line.

Still, I hope the Robosapien will stay around for a while.