|
|
Post by Dingo on Oct 8, 2005 11:35:47 GMT 10
|
|
|
|
Post by Robotman on Oct 13, 2005 11:12:01 GMT 10
oh yes. This makes me think why bother with big robots when a small one is just better for accidentally stepping on while it's cleaning the floor.  But seriously, I love it! And it gives me a good starting point for an attempt at making my own. Man, it just looks so good. And if Almir Heralic is the young lass in the movie, then she just looks so good too  Keep 'em coming Dingo! |
|
|
|
Post by donburch on Sept 21, 2006 0:13:00 GMT 10
Cool ! She claims that it only took 3 months and is low cost, so let's all make one !!!
I did note the tether cables, so all the actual processing is done on a PC - but at 30cm tall we can't fit everything into it ... like batteries or a brain. A reasonable trade-off; but why make it so small (requiring high precision machining and higher priced components) ?
Actually I am left wondering how much of it was true, and how much was showmanship (aka exaggeration or plain lying). For instance how long do you reckon the hardware would have taken to make from scratch ? Speech synthesis on a PC it trivial, and the vision tasks (light following, colour following and face following) are all variations on one theme and easily achieved on a PC with the likes of RoboRealm.
Why do I call it showmanship ? Partly because we only see the robot performing one task at a time; but mainly because of the "game" scene. It appears that HR-2 and Almir are conversing, but (a) HR-2 has no sound input, and (b) HR-2 executes 5 repetitions of each movement, whereas Almir only performed 4 and 3. My conclusion is that the movie was carefully scripted and timed to mislead the viewer into thinking that HR-2 is intelligent. And if one part was intended to mislead, why whould the other segments not also be exaggerated ?
But I still think HR-2 is cool (and I agree with Botman's assessment of Almir) and I'll be studying the documentation with an eye to estimating cost and effort to build a copy.
Update: Almir barely mentions HR-2's construction or costing, leading me to believe that she started with it fully assembled. The walking was hand programmed. Her documentation gives an overview of each discrete "experiment" at a level even I can understand ... and so I've saved a copy for future reference.
|
|
|
|
Post by Robotman on Sept 25, 2006 12:20:31 GMT 10
Almir does state that her father helped with the machining side of things. She may not have spent any money, but her Daddy (and probably the engineering firm he works at, or owns) certainly did. I noticed the tether cables too. I think she made it so small because of power-to-weight problems with the servo motors. As a concept it's great. With suitable high-density, multi-layer boards and surface mounted components all of the processing power can be packed on-board, but we don't have the batteries required to power the little thing (for long, that is). And yes, the whole thing is scripted. All of the robotics demonstrations (including Honda's running robot) are scripted. Put something in front of Asimo and he'd plow into it at full tilt. I agree with Jason Godesky (at www.anthropik.com/thirty) that Asimo is there only to impress investors - "Look at us, we are high tech and clever. We make super robot." I think that robots will continue to be mathematically programmed because the day we give them enough general problem solving ability is the day others will label exploitation of robots as slavery. A programmed machine (at least one that doesn't step on the cat) is ethically acceptable. A further problem was highlighted in a Doctor Who (Tom Baker) episode where robots on a mining ship started murdering people. Poor Lela (the Doctor's assistant at the time) freaks out because the robots look like dead people walking. The Doctor discusses how other societies that created "living" robots had to learn to cope with this, or else reject robots entirely. Very thought provoking. ... I would like to build a robot like HR-2. It will be bigger (but not man-size, maybe 400mm tall) and include stereo vision (it's not that out of reach to do) for basic collision avoidance, simple audio in (basic sound recognition - maybe not speech), phoneme speech out, balance sensors galore, goal driven self-learning to stand upright and walk (the "purpose of life" for this robot is simply to walk and recharge - see www.schursastrophotography.com/robotics/paami_main.html for a robot that two people have thought about this "life purpose" thing), arms with 4 finger-like grippers, and a cloth outer skin with capacitive sensors sewn in to detect touch. That's a lot, and a lot of planning, designing, resources acquistion, and money will be required. But I want to try it before the bloody oil runs out and our modern, agriculture based, high-tech society collapses. Rod |
|
|
|
Post by FSDSDF on Aug 14, 2008 17:50:38 GMT 10
|
|
|
|
Post by SFSD on Aug 14, 2008 17:52:01 GMT 10
|
|
|
|
Post by SDFSDF on Aug 14, 2008 17:53:43 GMT 10
|
|
|
|
Post by SDFDSF on Aug 14, 2008 17:55:16 GMT 10
I also know that when that decision is made you want to be on the road to a training program as soon as possible. free online games.Just remember that choosing a Training program for yourself and the dog is as important as choosing. play war games.a college. free online war games.You will need to know a lot about. online games.the program and what you really need from a specific Guide Dog school. |
|
