Burning Man 2009 Proposals

From OrbSWARM
Revision as of 01:41, 23 January 2009 by Christie (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

What do YOU want to make?

Burning Man 2009 Proposal Submission Deadline is February 1, 2009


Trilobite

Dillo it scuttles along

Self reconfiguring zen garden

Platform for shakespere

Lenticular images

Centrifuge hot tub

Lee's friend patrick

Doesn't someone in the East Bay have a hot-tub art car that spins?

UAV SWARM

A swarm of unmanned aerial vehicles Good resources: http://diydrones.com/ http://goflykite.com/ http://wildrc.com/ http://www.robotworldnews.com/100625.html http://heli.stanford.edu/

Pyrobots

portable SWARM-like fire platform... They could be pretty simple 4-wheel things with cordless drill motors and a modified-daughterboard controller. A rotating platform on top (1/8 aluminum?) with places to strap on pyro. With a rotating union to pipe propane from 2-lb bottles? Or rotate bottle a la Bob's fire chandelier? If it rotates fast enough could put POV led array for fire-less venues, like hypknowtron but smaller, more sophisticated. Also could use gnarly high-power LEDs for pyro simulation.

Basic electronics would be cheap and less-over-engineered than SWARM

swarming rocks

(from Charlie) 10 wheel driven robots that look like nice rocks (or something else that is comfortable to sit on) they have nice lights and make cool sounds. they have sounds during the day all kinds of things can trigger it, they also have blinky lights at night, once an hour the rocks kinda lift a little in the air and engage their wheels they they move around a bit and reconfugure into a new formation, sometimes a circle, sometimes a line, i am sure there a plenty of patterns the could create. then the sit back down and blink and flash and make music for an hour and then they move again. maybe they move from the fence to center camp over the week, maybe they just hang around in one place. I think it would work with what you have and help with what you want things to do in the future.

Self-assembling robots (or at least a step towards)

(This is the result of a brainstorming session with Christie, Niladri, Corey Fro, Rick Taylor with input from Denise.) Work done in the pre-proposal process is collected on tessellations page

visualization of top surface

This swarm represents all of humanity’s incapacity to operate independently. It illustrates the strength in the group, collective action. While maintaining free independent operation, the swarm comes together to create something larger than the sum of its parts. We’re proposing a new kind of swarming robots. They are 19 hexagonal units that by combining become collectively intelligent but are weaker when operating independently. They should be relatively small (about the size of the orbs or smaller) and low cost so that more can be built to better illustrate swarm behavior.

The basic shape of each robot is roughly a hexagon surface between 2 and 3 feet in diameter, with a large rim to allow for a solid yet flexible interconnect. A single drive wheel with the capacity for high maneuverability should propel these around.

The shape of these would not be a perfect hexagon. By changing the line of each edge of the hex to an S shape, we can create a figure that is attractive and facilitates alignment of the hexes. We’ve discussed using light in the resultant tongue and groove interlock for identification. We can run a light pipe along the top edge to illuminate and define the line.

The manner of locomotion would involve a single wheel drive utilizing the motor controller technology developed for the orbs. For balance and control, 3 additional casters will be set on springs on alternate vertices. This should enable us to leverage the existing motor controller design. The drive motor will probably have to be substantially larger, as these will unquestionably be heavier than the orbs.

The design of the upper surface is still quite fuzzy. For now we’re thinking the best top face would be an array of LEDs where the face has lenses in a shape that reflects the cross-sectional shape. Some elevation to this would be good so they could be seen from a distance as one approaches.

The majority of communication will be done through the zigbee interface. There's obviously a need for each node to identify its neighbors, so we're thinking light on the interconnect would not only add visual appeal when viewed from the side, but enable each node to identify the others with colors representing names or addresses. When connected, elections could be held to select a leader that determines motion for the group. (I'm thinking similar to a spanning tree root bridge election.)

The hexes have several different operational modes, patterned, directed or emergent. They can operate independently and after locking together in nondeterministic ways, as a single entity. They can communicate and interact with those in proximity or at a short distance.

In patterned mode, each hex can execute a predetermined set of instructions for shows and demonstrations of their maneuverability and control. This would provide our opportunity to show off any new gadgets or gizmos we come up with.

In the emergent mode, they’ll connect together organically. The interactions will not stop at connecting together. Once connected, they will communicate with their neighbor in such a way that organically creates mood and visuals and movement.

Directed mode would be how we get this/these to where we want them to play. It would allow us to make them appear to interact with things they cannot sense. Directed mode would require that we have a control console. We intend to achieve control by connecting the console in the same manner another hex would be connecting up. By having it always win elections for leader, it would be able to direct the movements of all the other hexes.

We’ve decided to take out the GPS Power supply(sic). While it would be helpful in running patterns, the added cost of including it would defeat the goal of making each individual hex as low cost as possible. We should actively examine and select all potential components based on their impact on the bottom line. Even small differences, once scaled across the number of modules we intend to make will add up quickly.

The biggest technical challenges I(Christie) forsee include alignment control for locking, transition from individual to block control and control decision communication while in block mode. I'm sure there are other pretty significant issues we'll face.

We came up with a partial parts list to get an idea of what it'd take to make each one. Although this is probably only useful as a starting point, it should give folks an idea of what we expect to put in them.

Per robot parts:
Motor controllers x2
Motors x2
Zigbee
RGB LEDs
CPU
Batteries
Sound Board
Speakers
Accelerometers
Drive wheel
Caster wheels x3
Acrylic Rod
Contact Switch
Frame*
Locking mechanism* (commercial electronic lock?)

I (jon) haven't seen anything like this outside a research lab http://www.suicidebots.com/2007/02/27/more-modular-robotics/ (no doubt because they are hard!)

More here: http://www.spatialrobots.com/archives/category/spatial-robots

Dirigible Snake

(from Lee) A long, multi-segmented snake-thing made with tubes filled with helium, propellers, contracting wires and the like. In calm winds it would be launched and squirm around in the sky, slithering here and there, freaking everybody out. Pointer/Cursor Kite inspiration

From Jon: tether it to the ground, and use a large blower like from bouncy-castle? Remote-control servo flap valves to different segments could inflate/deflate for interesting & unpredictable motions. Illuminate from within or without.


Emergent playa-evolved form (Dawn)

(This is cross posted on the Flaming Lotus Girls Wiki)

This year's theme is evolution. This, plus a few other emailed comments, has got me thinking about an idea. Bear with me as it hasn't fully formed, let's just say that it's evolving (heh). Theoretically, the beings we are evolved through time from some random fish who got tired of swimming in the ocean and decided to check out dry land. Thousands of years later, human beings as we are now have evolved into what we are, creating what we do. Some of us have evolved to the state where we trek out as a tribe to the desert, bringing our ideas and gizmos with us.

The playa has years of exposure to all sorts of random stuff: people, electronics, fire, paint, random chemicals, explosions, ideas, snow, ice, metal, electricity, etc. What if the next stage of evolution doesn't come directly from any beings currently in existence, but instead crawls directly out of the playa, the result of years of exposure to all the stuff I just mentioned and some spark to rise out of the playa?

So what would this being be? What would it look like? This parts a bit fuzzy, but what I'm thinking is that it evolves out of cracks in the playa over the course of the week. Cracks could be created out of the playa on frames sort of like chasm from 2004: http://www.horg.com/chasm/ only made to look more like something hatching or growing out of the playa. Each day it would emerge more clearly from the playa. My idea is sort of an enclosed pod, a sort of eggshell outside at first, but as it opens out this is actually part of the being itself. It's final stage of emergence would occur Friday or Saturday night.

Rick and I discussed a ideas for what sort of creature would emerge from the playa and thought of few things: it should have 3 arms and hands (because didn't you ever think you needed an extra hand for something?) with 6 fingers, the extra finger being a second opposable thumb on the other side of the hand (pinky side). We also thought that being out there exposed to the extremes of temperature and all that fire, that it must breathe fire. As it's emerging, for interactivity, it could sense people around and made noises, blink lights, etc. maybe slow down it's emergence from the playa.

It could either evolve as a self-contained, pre-built to emerge being, or as it evolves we could add to it over the course of the week. It would even be fun to have it interact with the orbs if they felt like showing up. ;)

I think this could either be really involved, or it could be done more simply, depending on interest, funding, time, materials, etc.

I welcome any input, suggestions, comments.