this is a project where I designed and built a work station to a six degree of freedom robotic arm. It is a common example from a production line, that the work piece is travelling on a conveyor belt or palette to the work station, where it stops in a fixed position, and the robot arm takes over the completion of the procedure from there. My main goal was to make experiments, if the robotic arm is enough precise to assembly work pieces during they are moving on a conveyor belt. This "on-the-go" assembly (or other industrial tasks) are not widely used in the industry yet, but they will be in the future.
How it works?
There is an incremental encoder on the lower side of the motor of the conveyor, which gives the information to the main processor unit of the robot arm, to be able to calculate the actual speed and displacement of the conveyor.
On the side of the conveyor, there are several inductive proximity sensors, which can detect the aluminium palette, when it pass by them. Using this information, the robotic arm gripper can follow the palette with the same speed, and it is able to do all the assembly work (or whatever task we gave to it). The speed of the conveyor belt is adjustable with two frequency inverters. The palette can be stopped in several points with pneumatic stopper, and it returns to its original position with the help of a pneumatic way selector.
I did not built all the parts of the conveyor belt, as a base I used parts of an old disassembled conveyor (which was used in a Nokia factory for procure speakers into eg. 3310 mobile phones :) ). The encoder, the palette and other parts I made manually, and of course, the programming, which was the hardest work! :)
You can see here the work station, during doing one simple assembly task with only 30% of the maximum speed.
I am now in the topic of 3D printing, and planning to buy a filament feeder, hot-end and nozzle (parts of a 3D plastic printer) and put it on the robot gripper head. In this way I can try out to print large objects (in the maximal size of ~ 1,2m * 0,8m). May it would be great to extend this head with a computer fan, so probably the plastic filament will cool down very fast, and just a very few substantiation will be needed under the printed object.
Of course this will be only an experiment too, because I can use my university's robot only in a limited time :)