News for hackers. Stuff that matters.
Re:Tesla unveils new car-charging robo-snake by Animats on Thursday August 06, 2015 @04:09PM (#10019082)
Nice. Robotic fueling has been done before [1][2], but this is much less clunky. This looks like the OC Robotics snake robot.[3] Finally, a use for snake robots, which have been around for 25 years but are not used much. Tesla has a good application for this - the car end is cooperative and standard. The car's parking guidance system can be programmed to recognize visual targets and get itself into the proper position.

The mechanism is simple. There are many linear actuators in the base pulling on cables that run through the snake segments and attach to plates at the joints.[4] Cable wear is a common problem, but that can be overcome. Charging robots won't cycle that fast; tens of cycles per hour, not thousands.

[1] http://fuelmatics.com/videos-2/video-demonstration/ [2] http://www.robosoft.com/robotic-solutions/transport--logisit... [3] http://www.ocrobotics.com/ [4] https://www.youtube.com/watch?v=WTkmyDO2ubs

Re:Tesla unveils new car-charging robo-snake by darkmighty on Thursday August 06, 2015 @09:51PM (#10020554)
Isn't this a little over-actuated for the problem at hand? It shouldn't need to go around complex barriers, just connect with the charging port. I don't see why something much simpler/cheaper with 3 or 4 degrees of freedom shouldn't be able to do the job. This is pretty cool though.
Re:Tesla unveils new car-charging robo-snake by nothis on Thursday August 06, 2015 @06:56PM (#10019980)
Call me a pessimist but I look at that thing and I see it breaking. Drive over it, vandalism, one of the hundred little motors having a defect... Why not just have two or three axis of movement in a more static, 90° construction??
Re:Tesla unveils new car-charging robo-snake by keerthiko on Thursday August 06, 2015 @07:08PM (#10020031)
The whole point of snake robotics is that one motor being a defect shouldn't render the entire robot useless. Of course that depends on intelligent control systems (algorithmic or mechanical) that can deal with limitations on any of the joints, without explicit damage sensors.

As for vandalism or getting driven over, well, that's not really a vulnerability exclusive to this approach.

Re:Tesla unveils new car-charging robo-snake by Animats on Thursday August 06, 2015 @07:21PM (#10020081)
For protection from vehicles, put a bollard on either side. If the cables have overload springs, which they should, forcing it to bend won't hurt anything.

The tentacle approach is mechanically elegant. The base mechanism is obviously a prototype, but that can be cleaned up for production. Ice, water, and sand in the mechanism will be a problem for outdoor installations, so the base unit has to be protected against those.

Re:Tesla unveils new car-charging robo-snake by justlurkin on Thursday August 06, 2015 @04:28PM (#10019197)
In my mind, I'm seeing a simpler solution that may be viable, enter the The "Claw Game" charger:

* Attaches to ceiling of structure

* Has some sort of sensor to identify charging port, perhaps on the charger end itself

* Has some limited range of mobility - 2'x2' with X & Y axis movement

* Cord unreels down to the charging port once it is reasonably lined up

* Charging cable has Macbook style electromagnet, reaches "charging zone" and activates, inserting itself

* Safety checks are done, wattage is pumped into the cottage if all is well

I honestly don't know if this is better or worse, it's definitely less cool.

Re:Tesla unveils new car-charging robo-snake by Animats on Thursday August 06, 2015 @05:09PM (#10019467)
That could probably be made to work today, with the gear used by container cranes to slot containers onto ships, trucks, and railroad cars. It would be more complicated than the snake, because stabilizing a hanging load is hard. Once in a while it would probably scratch the car body.

What a lot of people here aren't getting is that snake robots are mechanically simple. They're unusual, but neither complex nor expensive to manufacture.

Re:Tesla unveils new car-charging robo-snake by derefr on Thursday August 06, 2015 @05:44PM (#10019670)
Would the claw still work if it was coated in soft rubber?

Alternately, I'm imagining the cute-but-ridiculous solution of making the claw and the car magnetically repel one another, so as it the claw plumb-bobs downward and swings toward the car, it just sort of naturally veers away from the surface, rocking around it in an arc until it stabilizes.

Re:Tesla unveils new car-charging robo-snake by hexasquid on Friday August 07, 2015 @02:29AM (#10021159)
T-600s were covered in rubber - turned out people could spot them easily, so they were poor infiltrators.

Tesla should consider covering these with living tissue to make them more palatable to humans.

Re:Tesla unveils new car-charging robo-snake by rcthompson on Thursday August 06, 2015 @07:02PM (#10020013)
Make it so that the magnet is recessed inside a rim of soft plastic or rubber. Then it will still be able to slot onto the charge port but will not be able to hit the car body.
Re:Tesla unveils new car-charging robo-snake by Crito on Thursday August 06, 2015 @09:04PM (#10020427)
> Attaches to ceiling of structure

That's where my garage door goes.

Re:Tesla unveils new car-charging robo-snake by justlurkin on Friday August 14, 2015 @11:12AM (#10060529)
You're telling me you don't have a 1+ feet of clearance above your garage door?
Re:Tesla unveils new car-charging robo-snake by Patrick_Devine on Thursday August 06, 2015 @08:57PM (#10020409)
Some variation of this could potentially work with the Model S, or Model X, however the Model III will most likely have steel body panels instead of aluminum. I would imagine it would be next to impossible to line up the magnet correctly.