iRobot engineer builds stair-climbing robot for Masters thesis, not release
By Donald Melanson 
posted Apr 16th 2009 8:17PM
posted Apr 16th 2009 8:17PM
What does an iRobot engineer do for their Masters thesis? Build a robot, of course -- and, in the case of this one particular unnamed engineer, a stair-climbing robot. Unfortunately, any other details are a bit hard to come by, but it looks like the folks at iRobot were impressed enough with the bot to throw a short video of it up on the company's official YouTube channel. As you can see for yourself after the break, while it is a bit on the pokey side, it certainly seems to be more than capable of performing the task at hand with ease, and with some satisfactory buzzing and whirring sounds to boot.
I'm sure he's not really unnamed.
His/her parents didn't name her/him for fear that evil spirits would then be able to find him before he could liberate robots from their most foul enemy: stairs.
He has doomed us all!
It is cool and all but that would take forever to go up or down a flight of stairs, not very practical
Also here is a youtube video: http://www.youtube.com/watch?v=kqGOwqWBV-E
Oh oops the read link takes you to the video..
wow the same comment on youtube
Faster than regrowing legs so you can walk out of your wheelchair and go down them yourself.
im seeing for myself.... and there's nothing after the break......
iRobot makes good stuff .
I bought a Roomba and a Scooba, and while the Roomba has no alternatives I know of , other robotic vacuum cleaners are 3/4 times more expensive and just not as good.
I don't understand why they don't produce also a robotic lawn mower for a decent price, the actual models from Husquarna etc. cost upwards of 1500 euros and I'm sure that if iRobot can produce and sell the roomba for less than 500 euros (and practically pushed out of the market all other vacuuming robot makers and models) they will also be able to produce an affordable robotic mower.
Apparently this is still one of the commercial technology in which an american company leads the world, I whish I could buy more of their products and that they would renew their product line a bit more often.
Don't you think a lawn mower could be pretty dangerous ? I mean the cat may not appreciate the feeling of a blade on its tail
Robot lawnmowers usually do poorly in safety reviews, don't think they haven't thought about it though.
what they need is a robot with all sensors etc and just detachable part. have storage for grass or dusts.... (using dyson non bag technology) and detachable bottom ither have a tube sucking for vacumn or blade for cutting grass. (just dont forgot to change the blade over when bring it inside)
Finally! I can get my soda delivered without going downstairs.
I love the Microchip IDE box next to the Mcmaster catalog next to what I assume is a digikey catalog.... No Joke, that is exactly what is on my shelf in my research lab.. in that order... and exactly what should be in any robot builder's lab.
Reminds me of the iBOT wheelchair
http://en.wikipedia.org/wiki/IBOT
The iBOT is a very stable and mobile powered wheelchair developed by Dean Kamen in a partnership between DEKA and Johnson and Johnson's Independence Technology division. It is a medical technology, made to help people with severe mobility problems.
By rotating its two sets of powered wheels about each other, the iBOT can "walk" up and down stairs
Good concept. But can it pull an asimo? Didnt think so.
his prototype that he showed us at his master's defense was a little different, but it was really cool back over a year ago when he defended.
Sorry.. iBOT has this beat hardcore...
...and I drive past iRobot every day going to work ;-]
...just to explain a little more..
That process is so complex that I dont think you could do it with any speed. One step is fine but 15 or 30? ... how much time is that going to take? I think the current models at iRobot are faster at stair climbing...
The other snag is the depth and rise to the stairs... with this robots "crutch" foot maybe it can adjust for the size of the stair.. but I can see a very compact set of stairs or a spiral stair case being a big problem for this bot.
Its a great idea though... I'm wondering what his thesis was about.. I'm sure it wasn't titled "stair climbing robot"... that would have been shit we did in high school with legos...
I, for one, welcome our new robot Master thesis.
While this is fun to watch this isn't either climbing or descending in any type of "normal" approach. It is a best pushing itself up or in a controlled slide descending. It relies on leaning against the "nose" of the tread and the friction between the rubber crutch tips and the floor.
Real stair climbing/descending will only engage the top surface of each tread (and the landing). It is a complicated task that humans (plus dogs and cats) do well. From the first 2 or 3 steps our brain quickly appraises the rise and run of the treads such that we spend little additional bandwidth assessing the situation until we get to the top. Humans gain speed by intensionally "falling" (both up and down). Stairs with unusual rise/run ratios or changes in the ratio cause us to momentarily lose our balance while we reprogram. I expect that Asimo's famous stumble was a result of an error in rise/run assessment and the inability to quickly reassess before tumbling. Kamen's work is far ahead of Sony's as it negotiates changes in the COG.
Reminds me of my robot StairBot from year 2004.
For comparison reasons please have a look at
http://www.youtube.com/watch?v=ERrDB6qYdLw&hl=de
or
www.stairbot.de
Daleks handle stairs better than that does. At least the Daleks in the new Dr Who do.