Small Digging Robot

[vidam]

I'm trying to model a small robot that digs. I have an idea to start modeling the physics of digging such as the force to break earth, and the force against digging into the earth due to the hardness of the soil, and the friction. I need pointers of good places to start so I can model this in Matlab or another software modeling tool.

[vidam]

I think maybe my post was not clear. What I need to know if good reference books on the Science of Digging, Articles and Publications that I could read to learn about the art of digging for Dummies.

Thanks in advance,

[vidam]

Right after I posted this, I remembered something from my Robot Control course at CMU. It's called Force Control and it is often used in manipulator robotics. I need to spend some time reviewing before I do a mock-up of things in Matlab. I also remember a book writen on the Denavit-Hartenberg notation but cannot remember the title anymore. It was the text book they used in the class.

[Admin]

My understanding from your other posts is that your robot will be microsized?

If so, you won't be able to dig in a normal manner due to laws of scaling.

Look into boring by worms, as well as other small animals that dig (ants, crabs, etc.)

Basically, look at biology papers.

[vidam]

Right, the goal is to make the robot between a quarter-sized or even smaller dime-sized.
Do you have any documentation/web-pages about the "Laws of Scaling"? Are you stating that one cannot model kinematic / dynamic models at this size? I thought of looking at how crabs/crawfish/lobster dig. Crabs are very good at digging. They can slip under sand in a nanosecond when you try to pick them up.

[vidam]

http://crnano.typepad.com/crnblog/2004/08/scaling_lawsbac.html

[Admin]

Are you stating that one cannot model kinematic / dynamic models at this size?

You can, but you need to account for other physical properties that you normally could ignore in the macroscale.

For one, friction is a HUGE and significant force in the micro and nano scales. If you don't model friction correctly, you are wasting your time

Also, mass doesn't mean very much in the nano scales, and instead electrostatic and van der waals forces are dominate instead.

At micro and nano-scales, traditional motors and batteries become useless due to surface to volume ratios.

At macro scales, today's artificial muscles are useless, but become very powerful at micro and nano scales.

Many other things too . . . there are books on this if you look it up.

[vidam]

I don't really have time to research books on this topic. I've already calculated math for the required energy for the robot using macro-scale values for gravity, taking into account, and energy loss of the battery and conversion losses. Based on what you tell me I 've been wasting my time?

If you have a book recommendation in mind then I will consider buying it.

[vidam]

I bought a book, Introduction to Robotics Mechanics and Control by John Craig. It is the third edition. It was very expensive. I guess now I can return it to the store based on what you are telling me.

[vidam]

Doing an insect study for a possible design for species that can travel great distance and are earth - diggers as opposed to earth-dwellers.

Crabs - walk/swim for longer distances/sand/rock dwellers/loners (elicit no swarm-like behavior)

Ants- earth-diggers-dwellers/travel great distances to forage for food/swarm-based creatures (use pheremones)

Worms - earth-diggers/dwellers spend too much time under the earth as opposed to on top of it.

Cockroaches - walk/swim/survivors/travel great distances/dig-well?

Beetles- ?

Centipedes - earth-diggers-dwellers/too slow and don't travel well/digs well.

I'm not really up on my insects unfortunately. Any thoughts?

[vidam]

I think "The CrabRoach-Hopper" a cross-breed between a crab, grasshopper and a cockroach. I wonder if there are any insects like this in existence. Taking the ability of the Grasshoppers hopping ability combined with the design of a cockroaches to travel at amazingly high speeds with the crabs claws that help them cling and dig you've got the perfect insect-robot.

lol

[Admin]

I've already calculated math for the required energy for the robot using macro-scale values for gravity, taking into account, and energy loss of the battery and conversion losses. Based on what you tell me I 've been wasting my time?

If you don't know energy loss from friction, and friction is a significant value, then that means your energy calculations are wrong.

You will have to do research on microscale friction.

[vidam]

Thank for the tip to research micro=scale friction for micro-robots.


As a side note, maybe you should specify in your energy tutorial that it won't work for micro-robots unless you account for friction.

[Admin]

I've actually been thinking of doing a whole 'laws of scaling for robots' tutorial for awhile now . . . sooner or later someone will want to build a microrobot and need tips . . . you are the first for SoR

My energy calculator will work for any size robot. You just need to know how to enter in friction correctly for the Estimate Losses section. Calculating that loss for your robot is beyond my expertise . . .

[vidam]

I found this article if anyone is interested in it. If anyone wants the article just contact me.

Physics-Based Animation for Qualitative Assessment of Biomimetic Subterranean Burrowing Behaviors

Bergeron, Bryan 
HST Division of Harvard Medical School & MIT, 258 Harvard Street #315, Brookline MA 02446, USA. [email protected];

This paper appears in: Computational Intelligence in Robotics and Automation, 2007. CIRA 2007. International Symposium on
Publication Date: 20-23 June 2007
On page(s): 85-89
ISBN: 1-4244-0790-7
DOI: 10.1109/CIRA.2007.382837
Posted online: 2007-07-16 13:18:47.0
Abstract
Physics-based animations executing on 3D game engines enabled with physics middleware libraries and coprocessors can be used to explore the suitability of potential robot behaviors in working environments and robot configurations that are ill-defined or difficult and time-consuming to model with traditional quantitative tools. We use an inexpensive game development engine and PC hardware to develop physics-based animations of potential biomimetic subterranean robot burrowing behaviors. Qualitative assessment of energy efficiency, burrowing time, and digging capabilities of several biomimetic robot designs are validated with data from physical prototypes operated in a range of soil types and models of soil using colored particles. Results suggest this methodology is applicable to rapid screening of potential robot designs intended to operate in a variety of domains.l

[vidam]

Anyone heard of Ariel the crab-robot? It's not small as I would like but demonstrates the burrowing/digging behaviors.

http://www.berkeley.edu/news/media/releases/2002/09/rfull/robots.html

[neopj3]

depending on what kind of material you want to dig through (nothing smaller than large filter sand(get all the small stuff out)) or small fish rocks you could make the diging tool work for you as a locomotion tool as well.  think earth worm.  ever see the movie tremmers? a cone shaped nose and a set of small tank treads that run along the body would do the trick.  small pager motors and o-rings and a few other things macgyver would carry around and you've got the machanics.  now the question becomes how do you fit the brain in?  as said above muscle wires might work, a bit slow though.  if you said C cell battery size you might be able to fit the $30 micro in there.  my 2cents.  it can be done.  some of the worlds most important insects are the smallest and simplist.

[paulstreats]

I suppose that on a micro scale, you would just be brushing smaller particles or molecule clusters out of the way rather than actually digging, i imagine that at a micro type scale digging through something like soil would be similar to us digging through those small polystyrene balls in bean bags

[Gertlex]

The idea that comes to my mind is that while the dynamics and numbers will change as size changes, to a degree, the best digging method for a given material will be constant.  At some threshold, weaker effects such as those Admin mentioned will obviously necessitate a drastic change in design.

A similar example is using wheels on a "smooth" surface.  Obviously, when you get down to a small size, smooth surfaces are rather hard to come by... at this point you'd probably seek something better capable of handling varied terrain, whatever that method might be.

In this case, I don't think quarter-dime size goes beyond this size threshold.  At least with sand...

You mention crabs as an example.  Those generally dig in sand or sandy dirt... loose stuff.  If you're going for firmer ground, my gut intuition says that you would need a different design at that size threshold; I have a hard time picturing a crab digging (at all quickly) through a hard dirt layer, and clay.

Granted, various designs can handle different thresholds, too.

My ultimate point is that a reasonable approach might be to determine the digging mechanism first; testing several that seem like they would scale to your desired size, finding the best and thoroughly modeling it.  From there trial and error, perhaps.

I've got less experience here than you, I'm pretty sure... so what I've said above probably is how I'd approach the problem, with the partial goal of not overwhelming myself in initial research of concepts.

[ed1380]

I bought a book, Introduction to Robotics Mechanics and Control by John Craig. It is the third edition. It was very expensive. I guess now I can return it to the store based on what you are telling me.

I might not know all the physics and whatnot, but i do know there are ways of getting books for cheap

[vidam]

I've actually been thinking of doing a whole 'laws of scaling for robots' tutorial for awhile now . . . sooner or later someone will want to build a microrobot and need tips . . . you are the first for SoR

My energy calculator will work for any size robot. You just need to know how to enter in friction correctly for the Estimate Losses section. Calculating that loss for your robot is beyond my expertise . . .

For anyone interested in scaling in mobile robots there is a whole chapter on it in this guy's thesis I am planning to read:

http://asl.epfl.ch/aslInternalWeb/ASL/publications/uploadedFiles/Caprari_phd2753.pdf

[vidam]

I've actually been thinking of doing a whole 'laws of scaling for robots' tutorial for awhile now . . . sooner or later someone will want to build a microrobot and need tips . . . you are the first for SoR

My energy calculator will work for any size robot. You just need to know how to enter in friction correctly for the Estimate Losses section. Calculating that loss for your robot is beyond my expertise . . .

I would be interested to write a SoR tutorial for Laws of Scaling?

[airman00]

I've actually been thinking of doing a whole 'laws of scaling for robots' tutorial for awhile now . . . sooner or later someone will want to build a microrobot and need tips . . . you are the first for SoR

My energy calculator will work for any size robot. You just need to know how to enter in friction correctly for the Estimate Losses section. Calculating that loss for your robot is beyond my expertise . . .

I would be interested to write a SoR tutorial for Laws of Scaling?

PM admin for a member account

[gem_cat]

Mocap http://forum.unity3d.com/threads/what-do-you-think-about-cheap-royaltfree-mocap-services.187145/ offers a path forward. I am convinced you have a good idea. Find a creature you want to model and use the mocap to define its actions. My idea is to model an armadillo because it is very good at moving material, not just burrowing. I think you will find most digging animals use a technique similar to swimming.

[mklrobo]

In terms of power and scale, the only thing that I can think of, that is really small, are nano-bots.
At molecular size, they can be programmed, but what company does this, I do not know.
Power is going to be an issue, no matter what scale. Movement, burrowing, will be a job to do.
A Chemical engineering direction is an option to use to your advantage. Good Luck!

[GrimBot]

Have you figured anything out about this yet? I'm thinking of something similar. My thought on a digging helper robot is essentially an upright, mobile shovel with a weight on it.  It would position itself, put the shovel to the earth, drop a weight down the shaft to give the digging utility a bit of force, then I think it would hoist the weight back to the top of the shaft, maybe tilt a little to and fro to loose the ground it just sank into (at this point the weight would be at the top to give more leverage, but if it didn't sink it, the tilting may cause the bot to tip over from being top heavy and off balance). Then it would lift itself out of the ground and scoot forward a bit and do it all over. 

This ones idea is to just bust the dirt up a little, I would think that after a day or two, a good row of dirt would be broken up and be more manageable for the next robot, or the same one with a diff attachment.

Anyway, I hope that helps