Hello.

We are a new team, so now might be a good time to introduce ourselves, and our intended experimental robot, called "NetBot".

We are a bunch of programmers and Engineers who wanted to do something different – so NetBot is a lightweight class “NetCentric robot”.

In NetBot we do not intend to use a normal radio control. Instead, NetBot will have an onboard CPU, an onboard camera, and a 2.4 GHz link to a notebook computer (we had a brief chat to Brett about this.... his feeling was that as long as it is a legal frequency and does not interfere with anyone else, it is likely that nobody will mind what frequency we use in Australian events). The notebook computer will display the camera's robot-eye view, which we hope will make it easier to target weapons. Having a CPU on board should make it easier to program up autonomous behavior.

With NetBot we have pushed the software side, to produce an open infrastructure. The controller that controls NetBot can act as a server for multiple other notebook controllers and/or multiple robots. So multiple people can control different functions on one robot, or a single person can control a swarm of robots.

Also, NetBot can transmit pictures to any notebook (or any other type of computer) that is authenticated on the network.

Along the way there have been many false trails and problems. We have discovered that the camera and CPU need to be on a different circuit to the motors because when the motors draw power the voltage drops and the Operating System reboots! So we need a different circuit, with its own batteries. All this adds weight. So NetBot starts with a bit of a handicap. We are hoping that once we get the design sorted out, we can add autonomous behaviors to compensate.

Will it work? Well, it works in simulations, but the robot isn’t finished yet, so we don’t know.

Progress so far:
1. All necessary parts have been bought and weighed (chassis, motors, wheels, gearing, batteries, circuits and electronics come to about 26 kg. If the wiring weighs more than expected we will be drilling holes in the chassis).
2. The CAD is finished, metal cut and most of the chassis is welded.
3. The armour is started, but weeks from complete.
4. A Beta of the server-side software is done.
5. A Proof-Of-Concept of the Robot side software is done and communication with server tested.
6. Simulator built and "driving tests" run.
7. The circuits for the Robot-side hardware are designed and mostly in place, but one IC is weeks from completion.
8. Operating System installed on the Robot and mostly running, but camera does not work yet.

We hope to be ready by RoboWars II, but with less than 7 weeks left, we will need a small miracle. If not RoboWars II, maybe for the next battle

wow. thjats very impressive i must say and good luck with it.

There is an american robot called cyclonebot which is doing a similar thing and uses computers for conttroll and a gps unit for its tralational spinning movement.

HOpe it goes well for u guys. whats the weopon also.
_________________
Andrew Welch, Team Unconventional Robotics

Weapon? You want a weapon! No, seriously, if we succeed in getting netBot ready by RoboWars II we won't have time to mount anything beyond a simple ram. Later we should be able to do more.

The software currently supports the concept of self-describing weapons modules, so in theory you could change weapons between bouts, depending on who your next opponent is. Thesoftware reconfigures the user interface (on the notebook) to control the new weapon.

It is interesting that you mention GPS. The software interprets an NMEA string from a GPS - but I couldn't find a use for it in RoboWars. Maybe if we fought in very big arenas I will look up cyclonebot and see how they use a GPS.

I'm also developing a 2.4Ghz system
using the Liapac 2.4Ghz trancievers
which is based on nordic semiconductors
http://www.nvlsi.no/index.cfm?obj=product&act=display&pro=64

you are using some kind of 802.11b/g/etc?
know what frequency band it uses?

mine uses 1mhz chanells from about 2.40ghz upto 2.50 so if we can find a way to fit? it also transmits in <4ms bursts so we may be able to share in that way.


if you can keep the noise down you could use a switching power supply to replace your seperate reciever/controller battery.

what cpu etc are you using?

Cyclone bot is a "thwack bot" (ie whole robot spins on the spot) then they conrolled it with soft core FPGA and some custom DSP addons from altera so they could do translation while driving. They didnt use GPS, they used a hall effect sensor to pick up earths magnetic field then did DSP on it to get an accurate direction reference then modulated motor power synchronised to that based on controll inputs. Perhaps most important of all they had a strip of LED's on the side that they could use to flash taunting messages at their enemies while spinning.
Quite Funky

I'd be interested in hearing more about your setup feel free to PM me in some way (although yahoo dosent seem to be working atm)
_________________
Mechanical engineers build weapons, civil engineers build targets

Valen: you can email me at aeldric(remove this space) @optushome.com.au

Happy to chat about the technical and software side - it was quite interesting.

Further welding today. Welding nearly complete. Mounts for motors, batteries, and motherboard still need to be added - should be finished next week.

New programmed behaviors:
1. Go straight to 0 velocity.
2. Go straight to maximum velocity (as distinct from the usual behavior, which is to accelerate in a graduated fashion).
3. Spin on the spot at maximum angular velocity (spinning right).
4. Spin on the spot at maximum angular velocity (spinning left).

All programmed behaviors worked in simulator. Hopefully we will put the robot together in 3-4 weeks and find out if it works for real.

hmm almost missed this thread its always good to see the new entries that arise from robowars. seems like this one will be no different!

shame its a lightweight, any plans for a featherweight entry ?
_________________
www.demon50s.com - Minimoto parts
http://www.youtube.com/user/HyzerGlen - Videoooozzz

We would love to make a featherweight, but there are several problems:

1. The bot has two cameras on board (botCam and weaponCam). We intend for the bot to be able to do some limited "thinking" for itself (at the moment we haven't got that far - it is dumber than Anna Nicole on valium). Controling the robot and these cameras will require an Operating System (LINUX) with a motherboard, CPU, custom-made circuits, seperate wiring and battery. This stuff is expensive so we put armour around it, in the light-hearted hope that when you guys tear our beautiful bot to bits, all that expensive circuitry survives. All that stuff weighs a ton. It represents over 10% of a lightweight's weight allowance. It just wouldn't work to put it on a featherweight.

2. All that stuff is physically big. Even using a low-footprint motherboard, it is 19 cm x 17 cm x 7 cm. Again - a bit too big for a featherweight.

I hope we can find a way of making it small enough to put in a featherweight - maybe there are people here with ideas.
_________________
David C.
Team SubKor

I wonder what happens if I do this.......

Done in the last week:
- Frame welded. Motors and batteries bolted in.
- Camera installed and running.
- Networking installed and....err.... mostly running.
- Armour started.
- All items weighed. Total weight: 26.2 kg.

With luck, we might be finished in 2-3 weeks.
_________________
David C.
Team SubKor

I wonder what happens if I do this.......

Completed this week:

1. The controllers aren't fitted yet so we ran the robot without the controllers. Robot is fast Minor damage was inflicted to the house Wife not happy

2. Microcontroller programming almost complete. Should be complete today.

3. LINUX driver programming should be complete in the next 5-6 days.

4. Armor lay-up continues. Should be complete in 5-6 days.

5. We should have time to mount some kind of ram, so the robot will have at least a nominal weapon. One of the team will fabricate a ram this week, and the frame will be reinforced to support the ram.

Still on track. Competing at RoboWars II still a possibility.
_________________
David C.
Team SubKor

I wonder what happens if I do this.......

any possibilty of using the bots inbuilt camera to take a photo of itself to put on the forum :p

sounds likes its coming together nice and well, i dont think youll have a problem getting it ready in time, best of luck!
_________________
www.demon50s.com - Minimoto parts
http://www.youtube.com/user/HyzerGlen - Videoooozzz

OK.

We have some photos of the development process (mostly of the armour lay-up, so pretty boring). I will take some photographs of the "naked" robot later this week, then photos of the robot with some of the armour bolted on (we haven't finished all the armour yet).

I'll put it together into some kind of organised page and host it on my site - I can probably get it organised by next weekend.
_________________
David C.
Team SubKor

I wonder what happens if I do this.......

Some images of the naked netBot:

http://members.optusnet.com.au/aeldric/roboWar/images/netBot.jpg
http://members.optusnet.com.au/aeldric/roboWar/images/netBotSide.jpg
http://members.optusnet.com.au/aeldric/roboWar/images/netBotFront.jpg

And just out of interest, some of the forms we used for lay-up (mainly included because after looking at this picture, it occurs to us that it wouldn't be so hard to give this robot wings....):

http://members.optusnet.com.au/aeldric/roboWar/images/Picture078_14Nov04.jpg
_________________
David C.
Team SubKor

I wonder what happens if I do this.......

that looks so cool

use clear polycarb for armor.
_________________
Tim Team Reaper.

kewlness, looks like some good engineering has gone into that thing, sort of reminds me of juggerbot
_________________
www.demon50s.com - Minimoto parts
http://www.youtube.com/user/HyzerGlen - Videoooozzz