If you watch science fiction movies, the robots of the future look like us. The truth is, though, many tasks go better when robots don’t look like us. Sometimes they are unique to a particular job or sometimes it is useful to draw inspiration from something other than a human being. One professor at Johns Hopkins along with some students decided to look at spider crickets as an inspiration for a new breed of jumping robots.
Over eight months, the students studied the kinematics of how the crickets could jump up to 60 times their body length and land on their feet. Granted, 60 times their body length is only about 2.5 feet, but if they were human-sized that would equate to jumping across an entire US football field.
By using high-speed video cameras (up to 400 frames per second) to determine how the wingless crickets manage these huge jumps. They apparently use their limbs and antennae to stabilize themselves during flight. They also streamline their bodies to maximize their jump distance. The researchers have created computerized models that replicate the cricket’s jumping motions. They hope to use this knowledge and these models to design high-jumping robots to travel over rough terrain without the expense of building a fully flying robot.
We’ve seen some insect-inspired robots, of course. We even recall a Russian robotic cockroach. While we’ve seen a jumping robot called a sandflea, we didn’t think the way it jumped mimicked its namesake. You can see the crickets leap in the video below. Then find some shape memory alloy, start a Hackaday.io project and start building.
18 thoughts on “Robots And Crickets”
Wouldn’t their jump efficiency be related to its weigth? I mean, the scaling up may be limited somewhere. Interesting though is what can be learn from the study.
that is absolutely right.
when you consider that fleas pull 300+ Gs
The biophysics equations from a book I once read on the topic indicate a scaled up cricket would be able to jump about 2.5 feet if it were only redesigned to handle the increased loading. If you wanted a creature to improve its jumping abilities as it increases in size, it would need to be redesigned more dramatically.
Even if the reality differs in details, scaling effects are are definitely not intuitively linear.
I just remembered some details of why this is the case. Mass increases as a cubed factor of size increase, while muscle (and bone) strength increases as a squared factor of size increase (roughly the cross-sectional area of the muscle/bone). The consequence of this is that as a creature gets larger, it will have to have proportionally more muscle (and larger bones) to equal the absolute performance of the smaller version. To have improved absolute performance, the design needs to change.
the whole “robots in scienc fiction look like us” crap is only true if you watch/read crap!
Worse is when it is assumed they would think like us.
Not a fan of Asimov?
“You think robots care what some hack science fiction writer thinks? I killed Isaac Asimov on the way over here. Well, Isaac somebody.” – Bender Bending Rodriguez
“Listen up robot, if you’re gunna shoot, shoot to kill, warning shots are for pussies!”
The first time I saw the Honda robot walking on TV, I got a bit worried!
“You don’t understand, that’s all it does, it can’t be bought, can’t be reasoned with, can’t be stopped!”
My problem with Asimov, Heinlein, to a certain extent A.C. Clarke and few other of the “golden age” writers is they are so right wing!
And most of their stories have no women!
I’m a huge H. Beam Piper fan, in quite a few of his stories his characters see humanoid robots that have had function sacrificed form form.
Data from ST:TNG wasn’t a “robot”, he was Dr Sung’s attempt at an artificial life form.
The Star Wars ‘droids were mostly functional, C3PO was humanoid because he was a protocol droid, made for human interaction, the other droids like R2D2 and all the other droids were functional.
I still wonder though what those little black rover things were for, the one Chewbacca scares the solder out of.
Not always true. Some stories are about robots, and if they look like people in those, yeah, it’s bad. Some just *involve* robots, like Star Wars, and so who cares what the look like as long as it’s cool. Plus there’s stuff like Star Trek TNG that’s so far in the future that they look like us purely out of convenience.
That blinking light in slow motion scenes is terrible. And scaling up? http://www.memecreator.org/static/images/memes/2791433.jpg
The blinking light is due to the light source used for illumination. Most light sources flicker at frequencies unnoticeable to the human eye, but apparent in slow motion video. For purposes of this study it is a purely a cosmetic problem. There is no need to scale this up, since it is being considered for micro-robots
Another university PR material with lacking content.
Yah he is great at it, see http://hackaday.com/2015/12/10/graphene-super-caps-coming-soon/ and http://hackaday.com/2016/01/13/nanotech-makes-safer-lithium-batteries/
Both completely vapid articles. I got little out of the latter but the first was laughable.
This article is equally as vapid as the first. By my calculation no humans will ever leap a football field, they would be jelly inside the suit that allowed it with out some sci fantasy gravity field. Over a house yes but not 300 feet and how? easily.
It seems you just have to throw enough buzzwords into a research paper, be from some “prestigious” university, have pictures or an over produced video with some post graduates looking like vacuous drones in white coats posing with something insignificant and he will write an article about it.
Leaf hoppers are far more impressive anyway with their geared back legs.
Berkeley has a dozen publish insect inspired robots, and they are just glossed over. Several of them are even open source.
These east coast folks are great at citogenesis :\
Oh man! Trying to count up the number of insect-inspired robots, on either coast, is like trying to count the number of termites in a mound. What I mean is that your east-coast / west-coast rant is off base. Rodney Brooks? I don’t think he was a professor at Berkeley… (Although I’ll give you Mark Tilden for Tupac’s team.)
If your point is that the JHU promo video didn’t cite previous work in the field, or even acknowledge that it’s a latecomer in a 35-year-old robotics movement, I’d have to agree.
Please be kind and respectful to help make the comments section excellent. (Comment Policy)