PR2 Gets Better At Plugging Itself In
Writing by Evan Ackerman on Tuesday, 23 of February , 2010 at 3:58 am
When we first saw PR2 Alpha plugging itself into a standard wall socket last year, it was a sort of hit and miss behavior where the robot would stab at the outlet a bunch of times until it actually made a connection. As they say in the video, this is a “sub-optimal” behavior. I commented at the time that:
This is one of those situations where developing a robot that’s able to plug into a socket 20% of the time (but try over and over) is (I imagine) much more efficient than developing a robot that’s able to plug into a socket 100% of the time.
Now, I’m not sure whether I’m exactly wrong about my assertion… There are many significant hardware differences between PR2 Alpha and the PR2 in the above video, and undoubtedly a lot of programming was necessary for the one-shot plug-in. This stuff is a big investment of time and money. As a programmer myself, I totally understand how when something is sub-optimal, you want to optimize, and the improved behavior is certainly beneficial to PR2. However, my question is (still) whether the amount of resources invested in getting PR2 to hit that outlet 100% of the time is really worth it when a much lower percentage of success is still perfectly adequate when it comes to getting the robot to charge itself.
Either way, optimal is optimal, and the same library of behaviors that allow PR2 to plug itself in which such accuracy and precision will certainly come in handy when it tries to do other stuff, like tying my shoelaces for me ’cause I’m too lazy to bend over.
[ Willow Garage ]
Comments (4)
Category: Research
- Add this post to
- Del.icio.us -
- Digg
Comment by seabird
Made Tuesday, 23 of February , 2010 at 10:32 am
Gee, i dunno, I kinda like it when it struggles. It gives it more character xP Besides, this is painfully slow.
Comment by kwc
Made Tuesday, 23 of February , 2010 at 10:39 am
Thanks for the post. Your assertion is right — it guided us to success in Milestone 2. This was done by a small team to port our old Milestone 2 code to our new codebase and robot. While the increased accuracy was desired, I’d say that the new calibration, forearm cameras, and code base made it a natural consequence of updating the code. Your last sentence hints at some more motivation: this is a test for things to come.
Comment by Scott Stanford
Made Tuesday, 23 of February , 2010 at 11:26 pm
Also, note that plugging in right the first time instead of repeated stabbing is significantly better for the outlet itself. If your robot plugs itself in a few times a day, you’d crack the cover plate after a few weeks with the stab-and-retry method.
Comment by Alan
Made Thursday, 25 of February , 2010 at 10:44 am
I think an inductance sensor would be much faster. There has to be a degree of electricity which leaks through the air, and a wire loop and a bit of math could easily pinpoint the exact location the plug should be to optimally fit into the socket.
It seems like a lot of people go with cameras for show, because it seems like robots without cameras are blind, however if you want a robot to do something well give it the sense it needs for that task instead of something which is not as well suited.
Most humans actually use touch because the socket is at an awkward angle; we stab, and slide it across the surface with gentle pressure until it’s successfully in the socket, however a robot does not have this privilege because it does not own the socket and a user might be upset with the marring the robot might cause on the socket.
