DARPA Robotics Challenge: Team THOR

preview_player
Показать описание
THOR (Tactical Hazardous Operations Robot) was one of the humanoid robots we met at the DARPA Robotics Challenge, designed and built by students at UCLA and the University of Pennsylvania. We chat with Steve McGill of Team THOR to learn about the disaster relief scenario and how teams direct their robots in each part of the obstacle course.

Shot and edited by Joey Fameli

Tested is:

Thanks for watching!
Рекомендации по теме
Комментарии
Автор

I got excited when they started talking about FIRST robotics because I made my first short documentary on FRC!

Lukems
Автор

Been looking good in the button ups the past few videos. Its a good look on you. Yay robots

Xtafa
Автор

well somebody forgot to turn on the microphone :p

valentinotto_
Автор

I was involved in a FIRST competition recently. it was interesting, my team sort of got middle place (at the OKC regionals), so alas...

I have mixed feelings about something though.
a lot of the parts were a bit high-level (and expensive):
  cRIO / roboRIO, if programmed in C++, still basically glosses over everything;
ended up leaving most of the coding to another guy, figuring the experience would
benefit him more, but wrapper classes for everything does make it a bit easy.
whole program was maybe a few thousand lines at most...
  Talon modules also make things too easy (basically plug and play);
plug in the wires, not even doing ones' own crimp connectors for the signal wires.
did crimp connectors and soldering for the power wires though, but still basically PnP.
  ...
  and, these parts are expensive.

so, it ended up mostly about mechanics and hardware, with pretty much a lot of glossing on the software and electrical front. well, and needing to use some wire gauges that I suspect were maybe a little excessive (most of the required sizes were about 4 AWG or more thicker than what was really needed, and wire isn't cheap either).

though, luckily, wasn't paying for any of this personally, and it was interesting at least...


if someone uses more affordable parts (some perfboard and transistors, a Raspberry Pi, thinner wire, ...), they can do things for a fraction of the cost, and get a little more hands on (all the fun of writing their own scheduling code, determining whether the transistors can run the motor they want to use, ...)

though, at least the CIM motors were sane (cost competitive with other similar motors), so later bought a few of these for personal projects (though, still uncertain if I should have bought a more expensive 1.5kW motor though... but I wouldn't have really had enough battery for this, so alas, *, ... 2x 340W is probably good enough...).

though, granted, maybe there is a bit of a cheese factor of robots built mostly out out cardboard and hot-glue, and power-tool style motors and 3D printed plastic, with perfboard control electronics with most of the components heat-sinked with globs of clay.


*: only so many, so few, amps and amp-hours you can get out of a
lead-acid battery... (at least short of getting marine SLA batteries or
similar... but... expensive...).
this would be for moving a pretty good-sized load though (around 300 lbs or so).

BGBTech
Автор

Just imagine how far this tech could have advanced if the people with actual skill, knowledge and creativity were given access to the resources required. same goes for any of the sciences.

nunyabiznez
Автор

Been involved with FIRST for 13 years Robots = LIFE!! also GO FIRST team 829 Digital Goats!

wcxctrack
Автор

Rewtuser. Yep this was that same robot. I found it funny that they both "had the years needed in robotics" to make a robot look at a door and not know what to do :p

icebringer
Автор

Nice. The people engineering these things should see what real factories/power plants etc looks like in real life. They can be pretty damn labyrinthine. I mean really hard to get to certain places. Really really hard.

TheRealFOSFOR
Автор

0:23  "Many years of practice"... PFFT...  Right.

Spoif
Автор

its stupid that they need to be humanoid. there are far more efficient formats that they could follow.

space.cowboy
Автор

I'm pretty sure this was the robot that looked at the closed door and just gave up and derped backwards.

shockwavecity
Автор

I wonder how much this robot can Bench

MrRishik
Автор

I wonder how much the average cost to build one of these robots is

Cutleryloverlife
Автор

Even UCLA is in on this? I'm all for robotics. I just wish these people would think about where this money is coming from and what there creations could be used for in the future. Is it really worth it? Is this the only way to advance robotics?

invisibleaznDJ
Автор

Really like the interviews and all,  but what's with the over usage of 'so' lately?  It's put at the beginning of nearly every sentence.

MrStofkat
Автор

You be Thor and ill be Odin, you rodent, im omnipotent - But seriously though, calling it the Odin project would of been cooler imo :)

Ethan_and_Astra
Автор

I hear a noise canceling alithogram in use! Lol....

hubzcaps
Автор

finally somebody making human shaped robots to save lives!, now that is cheaper than pay a human for the life insurance if dies though

sergiokaminotanjo
Автор

Am I the only one that thinks 1:12 just looks like FNAF?

ipullstuffapart
Автор

Now give them swords and put them in an arena that I would pay to watch

ShadowBurn
welcome to shbcf.ru