Hylas Homepage

Mini Sumo Experiments

[First Mini Sumo]

First attack. The tyres were from LEGO, I've got different ones now ... The scoop looks nice. It's from an old ink jet printer ... The whole robot is still a bit too light for a Mini Sumo ... some lead weights maybe ...


Next in the row of Sumos was a 'Bot called "JoBot".
The chassis is made of aluminium. It consists of a single piece of metal which is folded to form. I have a layout which I can print out in variable sizes. The wheels are from Pololu (great stuff they got ...).
I created the original control board for an 80C32 ( the empty areas on the PCB are meant for the external RAM/EPROM ). Since my 80C32 wasn't fast enough ( when programmed in BASCOM Basic at least ) I changed it against an AT90S4414. They are nearly pin compatible.

[sumo side]

I have also used a DT107 by Dontronics.
You can see the sonar range module (SRF04, made by Devantech ).
The 'Bot was already programmable for "hunting".

[Line Sensors]

Attaching the CNY70 sensors for line detection to the scoop with epoxy glue ...
What I don't show here is how I had to peel them off again because of a little fluke :) .
The sensors work pretty well.

[new sumo bot]

One of my new MiniSumos. Here, I am using a different board which is based on Elektor's ATM18 project. The chip on board is an ATMEL ATmega88. Programming is mostly done in 'C'.

[different line sensors]

I am using different sensors here: these are TCRT1000 by Osram. They are attached to the scoop by adhesive tape used for putting up mirrors (great stuff :) ).
Right now I am having the problem of getting variable results from the sonar detectors depending on momentary motor load and remaining battery capacity. Lots of things to do ...
For debugging purposes I am using XBEE modules to send data to a pc.
The 'Bot's firmware (in 'C') features a state machine which does all the behaviour.

Mini Sumo fight.
Line Following.