Make a human size robot(5 feet)


build a humaniod robot that is 5 feet tall and can work.


also must be able to:



grab stuff


do common household stuff

to win: must send pic and video to me of your robot completeing taskes

Yeah! OK! Bring on the

Yeah!  OK!  Bring on the Robot Apocalypse why doncha?


I would have to take a second morgage to pay for it!

I do however have a 5 foot animatronic sketelton that I made that I use to talk to the kids on Halloween - The head has nice movement and it can talk either from an MP3 player or manual control.

Oddbot read my mind!

Ok, here’s the thing.

Ok, here’s the thing. generally speaking most of the members here don’t even have a rolling robot that is able to do those tasks (and I’m not talking about walking and sitting). I always say go one step at a time. Build a rolling robot. Add a gripper to grab objects. Add a speech recognition chip to be able to give it commands. Add a MP3 chip or a text to speech chip for talk back. Add vision. Program the robot to clean the room picking socks, toys, empty beer cans, whatever. Build a walking robot. Add balance sensors to it. Add speech chips. Add vision. Then program it to do house chores. If all works, go bigger. 

Just one point that everyone

Just one point that everyone has hinted at but not really said: You have to have built this bot first before you can make it a challenge.

I’d like $2 million but

I agree Chuck. As successful as Darpa have been I feel like they’re trivialising what it will take to make a humanoid capable of all this task.

Quite a Challenge

I love the Darpa stuff, but, it will be very interesting to see what anyone can do to touch this.  Of course Darpa may be happy with a telepresence (person driving the thing), which would still be a huge accomplishment.

I think humanoid robots capable of using human tools is a great idea, after all, most things are made to be used by people with human hands, human feet, human eyes, etc…  That said the effort level to get there is immense.

Some robot expert, (of course I forget who), said he could make a robot to fly a plane but not pick the keys out of your pocket.  Object recognition is the tall mountain we have yet to climb.

Think about just building a robot to empty garbage cans in an office at night.  If every garbage can is in a designated spot and marked with some robot recognizable mark, perhaps not too hard.  But if not specially marked it gets tough fast.  Even the dumbest person can pick out a garbage can based on its shape, its location, and what is inside.  Imagine trying to program the robot to do that.  I think a lot of files, luguage and parts bins would go out with the trash.

The human ability to parse imagery and match it against a huge database of objects is unchallenged so far.

Still it should be fun to see what the Darpa contestants come up with.  I could use 2 million dollars but somehow my SHR probably doesn’t …quite… qualify yet. …LOL.

Even that incredible robot

Even that incredible robot doesn’t meet this challenge.

Nice idea, but…

I thought the rules for bringing a challenge said that you had to be able to prove it could be done yourself.  I have been sifting through all of the robots here for a while and while some meet a few of the criteria, perhaps the challenge is either too ambitious, or perhaps a few too many steps ahead.  I mean, that cool Giger robot cost the guy a pretty penny to make, and it wasn’t five feet tall.  I would echo Oddbot’s statement about the first prize having to be kinda extravagant.

Points for enthusiasm, but maybe something a little more realistic.

Very nice concept, however…

You have a very unique concept for this challenge, however you do know that major companies, colleges, laboratories, and institutes use grants to pay for such work like this. If you were to build a life-size humanoid robot- take Asimo for example- you would be ending up somewhere in the premises between 1-2 million dollars. Anyways, I like the challenge though!

wow tthat is like a

wow tthat is like a movie…i hope you finish it succesfully… so i think the mayor trouble that you have is the programming part…there must tobe alot secuences and crossed suceunces, to think and take decisions… for example is diferent just walk and calculate distance than stay halt and grab a glass, but if the bot must to walk and grab a thing when it`s still moving…i dont know how to doit… the distances changes in every second… worst if the element to grab up is moving and the robot is stand up… i can´t wait to see it moving… or making anything…

Have you created one

Have you created one yourself?  I’d be curious to see how it works!


I wonder who want give so many money on a robot like that.

Need to be programed too.

Ask the NASA or Bill Gates to create it.