This is an idea for a light-weight spherical robot, which I propose to send.... to the Moon!

The robot consists of a small, weighty sphere within a larger, inflatable sphere. The small sphere contains all the robot's hardware - it's radio transceiver, power-pack, camera, and so on. The large sphere is simply a transparent beach-ball. I envision it being about six feet in diameter.

The small sphere is connected to the large sphere, and suspended within it, by six strings. These strings form three axes at right angles to each other, so that the small sphere is sitting at a sort of three dimensional cross-roads, with up and down in addition to left, right, front and back. Please feel free to take a moment to visualise this in your mind. Draw a picture if it helps.

Now, it should be obvious that by altering the length of some of the strings, we can move the small sphere around within the big sphere. If we shorten the up string, and lengthen the down string, the small sphere will just move a little closer to the top of the big sphere. If we move the small sphere to the left, the large sphere will be unbalanced, and will roll until the centre of gravity is once again directly over the point where it touches the ground. If we keep on moving the small sphere, by devious manipulation of the strings, the large sphere will keep on rolling. By this means it can be moved in any direction.

How is the length of the string altered? There are certain materials, commonly called artificial muscles, that contract when an electric current is passed through them. and expand again when the current is switched off.

I think that a robot like this would be very useful for lunar or martian exploration because, being inflatable, it would be quite light; it would also be rugged,and would bounce rather than break if it went over small drops; it cannot fall over, since it's sperical; and it's large diameter would allow it to roll easily and quickly over rough terrain. Keeping the camera pointing where you want it is the main problem; I have a suggestion for that if anyone's interested, but this idea is already quite long enough.
-- spacemoggy, Dec 19 2004

Some dicarded mars rover plans http://techreports....A-2003-tm212411.pdf
While the focus in this paper is wind power, it addresses many of the same issues I imagine your design would face. PDF [tiromancer, Dec 20 2004]

(?) Sphere-bot. http://www.egr.msu....tomikfil/page5.html
[2 fries shy of a happy meal, Dec 20 2004]

Remote controlled sphere, using pendulum drive http://private.pete....ru/rcl/iducoop.htm
[Ling, Dec 20 2004]

Another pendulum type, but 2 axis http://www.automati...ICRA_Cyclops_99.pdf
[Ling, Dec 20 2004]

spherical mouse spherical_20mouse
a similar thing, only the other way around. [neilp, Jan 10 2005]

This method IS old and even works on water http://www.jamesbon...-gadgets.php?id=007
Overall motion as center of mass is changed inside a surrounding sphere [Vernon, Jan 29 2006]

Ahhhh. I thought this was a vernon idea.

+
-- DesertFox, Dec 19 2004


This is really neat. Forget the whole 'moon explorer' bit and you've still got a totally new form of propulsion (someone will probably prove me wrong here). Also a very direct method of translating linear movement into rotational movement.
-- wagster, Dec 19 2004


I see a small problem: the Nitinol wires can only change length by a small amount, so the over balancing torque will by smaller than, say a straight forward pendulum drive.
However, I do have another idea, seeing as we are all thinking about balls.
-- Ling, Dec 20 2004


I don't see a light sphere having any traction over rubble and when on an incline it is trying to roll out of control. Six partially deflated spherical wheels on a vehicle maybe.
-- mensmaximus, Dec 20 2004


I like this, and not only because it reminds me of the spherical mouse.
-- neilp, Jan 10 2005


Baked, NASA's already working on that. I'll look for a link.
-- MikeOxbig, Jan 29 2006


Wow, this is way out there. It could work in theory, but it's definitely out of my range of thinking. Freaky.

[+]
-- TahuNuva, Jan 26 2008


Hey there. I came up with the same thing a few days ago so i started searching the net for other people who came up with the idea. I've come up with a few methods i'm trying to get plans for, liquid filled sphere, some kind of weight movement sphere with magnets, some kind of centrifuge method. Does anyone wanna work on this and build models in real life with me?
-- Taranoch, Jan 10 2009


am I the only person who thinks 4 wires would be adequate ?
-- FlyingToaster, Jan 10 2009


The offset mass (pendulum) method is still better. The string suspension system is going to be really complicated and fragile However they both suffer the same major problem with shifted mass sphere propulsion: You can't climb any kind of serious slope because even if you shift the mass to the maximum it still isn't over center relative to the slope. You have to add a secondary system for slope climbing or take a running start on every hill, and then your maximum ascent is limited by terminal velocity. And if you use solar energy for your power then you have a whole parcel of "walking with your face" type problems.
-- WcW, Jan 10 2009


I was thinking about this one. I think it would have problems with going up hills.
-- bungston, Apr 28 2010


Perhaps a hollow glass sphere, filled with water, with a goldfish?
A few plants in the bottom could photosynthesise to provide oxygen for the goldfish, and the sphere would roll forward as the goldfish swam around.
A small slug of plutonium would keep the water liquid through the lunar night.
-- coprocephalous, Apr 30 2010


[FT] //am I the only person who thinks 4 wires would be adequate //
Well, but maybe this lunar robot is intended to operate in a 5-dimensional universe?

Clever idea [+], though I'm not sure the proposed application (moon) is the best one for it. I see this more as an automated guard robot for capturing Number 6 when he attempts to escape from Portmeirion.
-- mouseposture, Apr 30 2010


I thought he'd moved to the Namibian desert?
-- coprocephalous, Apr 30 2010


Use six inextensible cords, and have a winch for each inside the core module.
-- notexactly, Apr 22 2018


If I'd been willing to wait two years and eight days, I wouldn't have had to have been looking for vacuum pump-related ideas, of which there are shockingly few, to have found this one.

Edit: No, actually, I don't think I found this one that way. I think I just looked at the ideas in the same category as the "Fluid-filled, Robotic, Ocean-Bottom Explorer" idea. But I stand by my assertion that we have too few ideas related to vacuum pumps.
-- notexactly, Apr 22 2018


Has paragraph breaks [+]
-- Voice, Apr 23 2018



random, halfbakery