DPRG List  

[DPRG] Hello again DPRG. (Re-Introduction) and long message.

Subject: [DPRG] Hello again DPRG. (Re-Introduction) and long message.
From: Quinones, Jose jquinones at ti.com
Date: Tue Apr 17 10:08:12 CDT 2012

Excellent topic!

You guys have gotten me thinking (again) as to why sometimes I look at robotics and it almost seems depressing. Don't take me wrong, the truth is that robotics have come quite a long way in the past 10 to 20 years, but I think the expectations were set a little bit too high (at least for me...). Think about the computer. 40 years ago (1970's), and how all we could do was blink LED's on the first personal computers whereas today we can strap to our belt a computer with cell phone, GPS, access to the internet, camera (for both pictures and video), game console, PDA, you name it!

When I was growing up, I remember drooling on a continuous basis as I browsed the HEATH catalog and their HERO line of robots. It was my expectation that by 2000 we would have butler like robots. At the rates computers grew, shouldn't that be the case? Well it is 2012 and still what you see mostly out there when it comes to robots are training platforms for kids and adults to learn about programming and robotics. Where are the life size bipeds ala Terminator? Sure, there are a few, mostly under research and development, but even then they are far from being what we thought your conventional Richie Boys would be able to buy in 2000.

I have also seen some quite complex platforms out there, but to deem them cost prohibitive is to think too positively (about 35K last time I checked, for just a platform with some sensors). And even then, there is really not that much they can do. For example, in these complex platforms I could make them carry drinks in a party and the robot would not collide with guests. But there is no way this robot could be made to actually make the drinks. In essence I would need to strap a second robot into the mobile robot platform if I wanted this kind of versatility. Could it then clean the dirty cups, take the garbage out, open and close the door as new guests show up, etc? Super simple tasks but it seems you need a single robot to perform each one of them.

The truth is that robotics are neither simple, nor complex. What they are is life! That is what we are trying to accomplish here. We are the first life form trying to create life (albeit non biological, in this case) and even for that there must be a learning curve. We are a brain trying to create a brain, or at least a fair portion of it.

I think that at the end, the successful robot will not be made of a collection of behavioral algorithms, but a core of algorithms that allow him/her to gather information and process in a way in which human-like learning can be achieved. At the moment, there might be computers getting close to doing this, but they take off entire rooms. Not to worry! The same happened about 100 years ago and we can put such computer about a dozen times in the palm of our hands, not to mention power it with batteries. The singularity is near!

Best regards,

Jose I Quinones

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://list.dprg.org/pipermail/dprglist/attachments/20120417/7e5f2841/attachment.html 

More information about the DPRG mailing list