A really smart ant

Discussions around autonomous, semi-automatic and intelligent robots and systems.

Moderators: BeligerAnt, petec, administrator

User avatar
Lincoln
Posts: 128
Joined: Sun Feb 29, 2004 12:00 am
Location: Olney, Milton keynes

A really smart ant

Post by Lincoln »

Just to let you all know that we are aiming to participate in this too, good to see how many others have been discussing autonomous - using standard (and therefore proven) sumo rules to start with seems like a good idea.

We're just about to send our PCB designs away to be made on PCBTrain. Using a PIC18F2680 for nouse and a mix of IR line detectors, proximity detector array (front and back) and a scanning range sensor.

Also included a gyro and 3-axis accelerometer - but they are more for trying to make a balancing robot than an autonomous sumo.

Not sure when completion is going to be though...

--
Christopher
User avatar
BeligerAnt
Posts: 1872
Joined: Wed May 15, 2002 12:00 am
Location: Brighton
Contact:

Post by BeligerAnt »

Chris, it sounds like you're taking this far too seriously! :)

Since we're talking processors, I am currently laying out an Atmel ATmega8 board which I will etch myself. I have already made some IR reflective sensors for line sensing (front and rear), and some of the Sharp OPIC sensors for opponent detection. No range detection planned for now.
Drive will be a couple of standard servos, run from 2 x Li-ion cells.
Gary, Team BeligerAnt
User avatar
peterwaller
Posts: 3213
Joined: Fri Feb 15, 2002 12:00 am
Location: Aylesbury Bucks
Contact:

Post by peterwaller »

Actually got the prototype moving last night at about 1/5 speed. I fitted two SY-CR102 IR send recieve units from Maplin at the front corners set to reverse the opposite drive motor while it is detecting the edge as a first try. The result gave edge avoidance but having once turned it follows the edge of the white line around the circle forever. Just need to increase the amount of turn then I think I will start on the opponent detection.
User avatar
Lincoln
Posts: 128
Joined: Sun Feb 29, 2004 12:00 am
Location: Olney, Milton keynes

Progress

Post by Lincoln »

PCBs are back from PCBTrain and partially populated. Using PIC18F2680 for processor. So far I've got control of a servo being driven +/-90degrees to scan - aim to mount a Sharp GP2D120 on the servo horn to give a search capability.

Having problems with IS471 proximity detectors - seem to get output locked either on or off - think this is down to IR light leakage from the emitters onto the sensor direct swamping the signal.

ST 3 axis accelerometer and Murata gyro calibrated and working, combinging their inputs with a Kalman filter to derive the tilt angle - not sure if I'm going to get it to work as a balancing robot or autonomous sumo first. I need to establish some comms with it now to assist with development.

I hope everyone else is having fun and making good progress too.
User avatar
peterwaller
Posts: 3213
Joined: Fri Feb 15, 2002 12:00 am
Location: Aylesbury Bucks
Contact:

Post by peterwaller »

I have now got the whole thing up and running.
It can detect the red cap off a WD40 can at about 40 cms.
At the moment it scans by rotating until it locates a target then pushes it out of the arena.
If it comes in contact with the edge it scans side to side from one edge detector to the other looking for an object.
At the moment if the object is out of range it doesn't go looking so I need to address that.
This is all on a vero board protoype but I have started building the final unit based on a tracked chassis using four of the 50:1 Inertia lab Sanyo like motor gearboxes and a suface mount PCB which is at the layout stage.
Adam Hargreaves
Posts: 223
Joined: Sat Apr 30, 2005 9:25 am
Location: Loughborough
Contact:

Post by Adam Hargreaves »

The ATR is nearly done. Not sure if it works. No line detection yet-no time for that (deadline next Fri) but I THINK that the opponent detection works.

Case looks good though-got it vac formed yday, and the graphics look nice. Watch for video or photos v soon
Team 'In Theory'
Andrew_Hibberd
Posts: 1134
Joined: Tue Jan 20, 2004 12:00 am
Location: London
Contact:

Post by Andrew_Hibberd »

loads of them 8)

need to replace the motors on mine, as the ones on there atm are not powerul enough to turn the tracks. Need to program it into a fight mode, which shouldn't be that hard as i already have avoid. Don't think i will get line detection/edge detection, so may not be very effective. I want to build something with more detail over summer so maybe aws21.
TEAM GEEK!
User avatar
Lincoln
Posts: 128
Joined: Sun Feb 29, 2004 12:00 am
Location: Olney, Milton keynes

Post by Lincoln »

well it's almost done :D it's sort of working. Just need to stop it requireing a bit of help
User avatar
BeligerAnt
Posts: 1872
Joined: Wed May 15, 2002 12:00 am
Location: Brighton
Contact:

Post by BeligerAnt »

Woo hoo! I now have a working microcontroller board that can drive my motors and detect black/white surfaces. :D

So now *all* I need is some software to connect it all together, some wheels, a chassis, body, armour, oh and some opponent detectors - and some kind of fighting algorithm! (All by AWS20...)

BTW I can thoroughly recommend the WinAVR development system - it's easy to use and FREE!
Gary, Team BeligerAnt
User avatar
petec
Posts: 555
Joined: Wed Jan 23, 2002 12:00 am
Location: Burgess Hill, West Sussex
Contact:

Post by petec »

*sigh* Wish I had spare time...... :cry:
------------
RobotWars101.org
RobotWars101.co.uk
Antweights.com
Antweights.co.uk
AntweightWorldSeries.com
Post Reply