My intention for Wall-E is to operate in two different modes. The first is a puppet mode for use when demonstrating Wall-E or where discrete interaction is required, the second is an autonomous mode where Wall-E does what he wants on his own.

To facilitate these two different control systems on one platform I decided to use a distributed control method that splits up the higher functions and the lower functions in much the same way as our own brains are meant to be working (does anyone really know?). This means that items such a motor control and head movement are not done directly by the main controller (brain) but by instruction to a lower level controller. For this the I2C bus (network) makes sense.

TWI two wire interface

Read more: Internal network design

So where do we start our Wall-E build? I decided the best thing here would be to start at what I saw as the most complex part, the track system. Joining the Wall-E Builders club and watching a YouTube video from tested.com showed me that it was possible but most people took an off-the-shelf approach to the track system.

Read more: Designing the track system - Pt1

When I decided to go back into robotics I wanted to build a platform for experimenting with autonomous robots. However I didn't just want a box on wheels or a couple of servos on a circuit board, I wanted something people could relate to. I made a short list of 3 which I would put to my wife and see what she thought.

First up we have Johnny 5 from the Short Circuit films and as my favorite robot of all time he made the list first.

j5

Read more: Why build Wall-E