Trial and error. More so error.

Yesterday was truly a test of everyone's endurance and patience. Given kits of Lego Mindstorm to go forth and construct a robot which would go forth and become our chess piece, we set about excited, reminiscent of childhood days of playing with Lego and the like. As the hoard of colourful and exciting pieces of all shapes and sizes, all hands on deck we got to constructing, with only minor hiccups when we couldn't find the odd piece or where it was mean to slot it. Our creation slowly took shape but when it came time to program it, things got a little difficult.

It was again like the protocol challenge, where we were given the starting point and the destination but our established protocol was challenged by the obstacles created by the walls so today the challenge was to program our robot to reach start to finish moving by protocol. Given a myriad of sensors we could program, my initial idea was to direct ours by sound. That is, in keeping with our Jester character, it would move forward two units and then diagonal one unit. The diagonal direction would be decided activating the sound sensor, moving towards the direction from which it hears noise which would be us clapping, like an audience entertained by the jester.

This proved to be a little difficult to measure the amount of volume which would trigger the sensor so we then decided to map out the journey. After some measuring and calculating, we used the robot itself to measure and record which direction to turn to then be programmed. This again caused difficulties as the path was influenced by many variables, the main one of which was the deteriorating battery life. After many attempts at reprogramming and remapping, we ended up as far as halfway down the hall.

It was a frustrating activity and we wouldn't have lost as much enthusiasm and motivation to keep trying if we hadn't had so much variation in our trials. If we were to keep going, I'd go back to the drawing board and program it's navigation as controlled by the sensors and now after having thought about it, like when we were the subject moving from point A to B by protocol, we still had the ability to make decisions based on judgment calls as decided based on our senses. We too were subject to variables when we made our journey, such as the one hour time limit, the impending rain and the fact that our batteries were perhaps a little flat too. The only way to overcome these variables is with judgment calls which is what we should have programed our robot to do.

Though it was a frustrating project which left us all completely exhausted and perhaps a little short with each other and stepping on some toes, what we achieved is a very steep learning curve in programing. I haven't had any experience in programming before and so what I picked up on was that you have to take into account all variables and be very specific in what actions you program as the smallest tweak can make a huge difference. By programming an element of decision making, it saves having to map out every single 'what if' that could occur along the way.

Onto the intertubes to look at what others have been creating with the same myriad of parts, I looked at discussions of what can be something different to create. Like many of the conversations we overheard and perhaps had ourselves, we joked about our robots smashing through the walls, doing our homework, programming it to program itself, teleporting and other such fantasties. Though these do seem very farfatched and perhaps not completely feasible with the pieces in the kit, it is that same sort of idea of the ability to make judgement calls and think for itself which is what I think is an element of what drives creation, which brings us back to creativity. To achieve new things, we have to think big regardless how ridiculous it might seem at first. Thinking about what else I could program it do, by taking advantage of the sensors, I would want to make it so that the robots interact with each other and move as such, like a dance, but so that they actively seek out each other for interaction. Much like, in fact, humans,

It didn't take this follow robot long to catch my eye so I leave you with a robot that has achieved what for me has only ever resulted in a square hole in the wall. I watched amazed last year as a group of people from my form class every morning would sit and solve rubix cubes. By programming the light sensor to read colour and precalculate the moves, it is able to solve the rubix cube in 60 faceturns.

Now I'm impressed.