Jump to content

Recommended Posts

Posted

Awesome, isn't it? That's a direct outgrowth of what I do - biomechanics and animal locomotion. In fact, I'm going to be applying to the lab of the prof who helped them with it for my PhD work (so if I'm lucky, I might be there in fall of 08).

 

The irony is that very little has been done in stablization and disturbed motion in vertebrates, yet the robot seems to be doing fine.

 

Mokele

Posted

That's spectacular. It's easy to make a machine that walks, but one that adapts its gait like that must take some significant processing power.

Posted

we got rid of the horse by mechanising the cariage to make the horseless cariage, and now we`re mechanising the horse too.

Fantastic!

the bit at the end with the roller skates astonished me the most, but that buzzing noise would have to go!

Posted

I was hoping to see how it coped with a sharp edge like a cliff or even a high curb. Still, very fascinating. I wonder how much bigger / smaller it will get eventually.

Posted

This is very incredible, but what are the practical uses other then entertainment?

 

I mean granted I'd pay $100 to try and kick and robot over, but I'm sure that if I managed to kick it over I'd do more then $100 in damage.

 

What I saw wasn't all that fantasic, the terrain it was moving over wasn't anything that tracks or treads couldn't already move across, not to mention it's stabilty issues and obvious weight restrictions.

 

It seems it's nothing more then a childhood obsession to build a Transformer or something else bipedal. Seriously though, how will this invention better mankind?

Posted
but what are the practical uses other then entertainment?

 

Seriously though, how will this invention better mankind?

 

in all honesty, if you have to ask that question, you Really wouldn`t understand the answer!

Posted

This is a nice achievement and it only is one of the first humble steps in developing machines, which perform tasks for us in bad environments. Of course, this robot is not as good as whatever animal with four legs (e.g. dogs, horses, goats), but still, it is impressive. Over time they will improve, and I'm quite confident that after ten years or so, robots will outperform real 4-legged animals in climbing rough terrains, keeping themselves upright under bad and varying conditions and with heavy and varying load.

 

Will it be of benefit to mankind? Depends on how it is used...

Posted

What I saw wasn't all that fantasic, the terrain it was moving over wasn't anything that tracks or treads couldn't already move across, not to mention it's stabilty issues and obvious weight restrictions.

 

Seriously?!

 

Its a massive operation getting that thing to walk forward let alone manage different terrains and immidiately correct itself after a massive kick off course - which it doesn't get any warning of.!

 

Animals at least can see whats ahead, this think probably has to judge its terrain every time it moves its front foot.

 

When its pushed it probably has to judge its angle of decent and place its foot at the correct distace away from its body, this thing in short is genius!

 

and if we can't find a scientific use? well think of the great fun drunk students of the future will have tipping over robotic cows!

Posted
It's easy to make a machine that walks, but one that adapts its gait like that must take some significant processing power.

 

First a small technical point: 'gait' technically refers to the pattern of footfalls, their duration within the cycle, and the mode of energy conservation used. The machine in the video was always using a "trot", characterized by the simultaneous ground contact of diagonally opposite limbs.

 

Also, it may not necessarily take much processing power; just use gyroscopes and accelerometers to figure out where the body is going and how fast, and put the legs in a position to absorb the force.

 

There was actually a talk just last SICB about "WHEGS", a little robot who could traverse amazing obstacles just by having wheels with large protruding spokes at 180 degrees (like legs) on a flexible axle. Apparently this is literally what roaches do: if the leg hits an obstruction, the apodeme (insect version of a tendon) keeps storing energy until there's enough to finally shove the animal past the obstruction. A springy axle, with no active control, gives it more manuverability that almost any modern robot of its size.

 

I was hoping to see how it coped with a sharp edge like a cliff or even a high curb. Still, very fascinating. I wonder how much bigger / smaller it will get eventually.

 

Cliffs are probably coming; the first studies on how animals deal with such things are only just beginning.

 

As for size, the Biewener lab (who helped on it, and where I'm applying in a bit over a year) is very big on scaling, because it has tremendous impacts on locomotion. Likely, it won't get any bigger (bigger than about 300kg and animals have to start curtailing their locomotor performance to stay within the capacity of bones and muscles without damage), but how much smaller is mostly limited by the ability of the engineers to make parts smaller.

 

What I saw wasn't all that fantasic, the terrain it was moving over wasn't anything that tracks or treads couldn't already move across, not to mention it's stabilty issues and obvious weight restrictions.

 

It seems it's nothing more then a childhood obsession to build a Transformer or something else bipedal. Seriously though, how will this invention better mankind?

 

Simply? Because wheels suck.

 

No, I'm being serious. Look at what a basic wheeled vehicle can do. Sure, it can go fast over level, smooth terrarin, but the rougher the terrain, the worse it performs. Steep inclines are impossible without special modifications. Gaps cannot be traversed except by jumps which require a running start (which again runs into the problem of needing smooth terrain). Water is impossible to traverse; wheels generate no thrust without special modifications. A jump from standing, again, impossible. Stabilization relies soley on a low center of mass and a few springs, and can easily become a problem is the vehicle is high-bodied. Landing from jumps and such also relies solely on additional components a vehicle may or may not have (shocks). Pure sideways movement is impossible without amazingly flexible axles, and mud is a *huge* problem, as is any slick surface such as ice.

 

Now look at a simple limbed animal, such as a salamander or cockroach. They can traverse very rough terrain with minimal loss of speed. Steep and even vertical inclines are possible, as is climbing trees, poles, etc. Jumps can be perfromed from standing, or the animal can simple reach a limb across. Swimming is easy, and requires no special modifications (dog paddle). Stabilization is active, simple, and relies on nothing not already present in the limbs for propulsion. Sideways movement is easily possible, mud is much less of a problem, and slick surfaces like ice are much easier.

 

Wheels are great in civilization, but require a special infrastructure of flat, smooth surfaces. Legs are much, much more versatile.

 

Mokele

Posted

I have to say that whilst I still have that annoying motor sound in my head, that was a very impressive video. The way in which its legs adjust to being kicked and over rough terrain etc. are very clever.

 

Whilst I can't say that I'm up to date on the latest advances in robotics, that video shows quite how advanced our robotics have become, and certainly shows a lot of potential for the future.

Posted

awesome!

i didn't see much running though, it never really modified it's style. i'd be really impressed if they could get the thing to leave the ground with every stride.

the honda biped has gotten to that stage, small strides but it's still a run.

Posted

It seems a bit clumsy, but if you compare it to two human beings carrying a large weight and not seeing where they're going, it's not bad at all (other than the incessant hopping). Compared to a cheetah, well ...

 

Nice work, still, it looks very familiar from the two-wheel balancing robot. As for pushing it, legs are way better than wheels. Eerie though, as mechanical as it looks hopping, when pushed it behaves strikingly like a live animal. That last push in the lab looked real spooky (I own 3 dogs that push each other around).

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.