Search This Blog

Monday, 21 March 2011

Robox Ep3-First Development (Ball Rolling)

We had an idea, now we had to figure out how to make it happen.  If we had consistently learned anything from our previous assignments it was the importance of thorough planning.

We were going to create a robot that randomly generated a path and printed designs on two surfaces, one above and one below, essentially sandwiching the robot between the two.  On the top we originally wanted to use floods of paint to produce a wash of colour.  We would fix a ball to the top of our unit which would 'roll' over the paint surface creating a trail.  Several problems arose: 1. We had no idea what to depict on the base, as the top canvas demonstrated the concept of duality which we were initially aiming to do.  The second canvas was somewhat redundant.  2.  On a much simpler note, gravity would take its toll on the paint and taint the bottom surface ruining the functionality of the design.  We changed to the idea of distributing the paint onto the bottom surface whilst using a large vivid tightly fixed to the top of the robot, to trace his path on the upper canvas.  We had a list of things to research.

Eventually after much youtube deliberation we decided the NXT hardware didn't provide the means to carry and apply the paint without help from some other system.  We looked into options (even considering water balloons again) ultimately settling on some sort of syringe, if we could make it work as an affordable and efficient option.

ignore his voice, or just mute the videos. Their pretty self-explanatory

Teammate Jason had been in the A&E at Auckland General week prior for concussion and suggested a trip back to flirt and weasel the nurses into letting us walk away with medicinal drip feeds which he assured us would be perfect for the task at hand.  I remember hearing stories of 'scarfie' med-students down south using drips to avoid hangovers and instantly doubted our chances.  We left with three (thankfully unused) syringes and feeds absolutely free.  Good thing the country's not enduring the aftermath of a great catastrophe at the moment and thus delivering frequent blood transfusions; Oh wait...  Kudos to the other lads for pulling it off.  Hope you got their phone numbers too..

It was also decided at this stage that we would use the idea of a sturdy wooden box surrounding the canvas area, this would both support the lid, and provide a reference point for the robot to bounce off (we would have to figure out at some stage in the building and programing of the unit how we were going to make him navigate the given area).

Most noticably we evolved the box design to separate the top and bottom of the box from the sides to allow easy transport, and access to both surfaces on a flat plane.  In turn, and with some advice from James we would also find a way to lock these when necessary. Other ideas that were scratched in favor of more efficient alternatives include using OHP paper on the upper canvas as opposed to a lid, to easily layer the two images. Also, using fabric sides for the box itself to easily see and record what the robot was doing during proceedings.

We now had functional designs and a very informal plan for how the whole thing was going to operate.  It was now time for the task of creating all the individual components.  Any more developments would be made as necessary, when necessary.

No comments:

Post a Comment