We also split this design into 3 parts

Paint Delivery

What it was:
Fig 1. Paint Delivery System

We designed a spray system that was meant to be less costly than spray pumps in the market. Our design involved using a peristaltic pump to transfer paint from the gallon, up the hose and into a secondary container. A modified handheld paint sprayer would then suck up the paint from the secondary container, and pump it out at high pressures through a nozzle, atomising the paint. We intended to modify the handheld paint sprayer by removing the need for a trigger, which was used as a “switch” to supply power to the pump in the sprayer, by connecting the pump to an Arduino instead. We also intended to modify its container to allow for a tube to be connected, in order to supply paint to the secondary container. 

Vertical Movement Mechanism

What it was:

 

Fig 2: Telescopic Rail System
Fig. 3 Our Vertical Movement System

We intended to have an extendable “arm/vertical track” similar to a telescopic rail system. The arms would extend using a pulley system as shown in fig 2. The central platform (where the spray would be) connected to the innermost arms would be able to move by using a timing belt pulley system.

Why we changed this: 

Stability was an issue, and multiple motors had to be used for the arm extension movement as well as the central platform movement, which would not be energy efficient.

Base Mechanism

What IT WAS:
Fig. 4 Arrangement Of Wheels In Our Base
Fig 5. Idle castor wheel attachment. The wheel will be able to move up and down as shown by the arrow.

We intended for our robot to have 4 wheels for stability. We wanted to achieve 3 motions with our base, which was to move right and left, move back and forward, and rotate. Our first design allowed for this to be achieved with only 3 motors, instead of four (i.e. 1 motor attached to each wheel). Each pair of omni wheels (each pair being the two wheels positioned diagonally from each other) would be connected together and given the same direction of rotation by one DC motor. This would allow for linear movement in the x and y direction, using two motors. For the rotation of the robot, there would be an extra idle castor wheel that can be raised and lowered by a third motor. When the idle wheel is lowered, it replaces both back omni wheels in touch with the ground so that the robot can rotate by making the motors spin in similar directions without fear of hindering from the on-axis wheels which are now floating. 

Why we changed  it: 

The parts needed to achieve this were not purchasable in the market. In order to reach the precision needed for this to work effectively, 3D printing would have to be used. However given our high estimated load of around 20kg, the 3D printing plastic material would not be able to hold up the weight of the rest of the robot. Hence we decided to explore other ideas.