Throwing down the gauntlet to objects

This post will detail more about the IR sensor mounting and the final testing we did with it. As we saw in the last post:

 Hardware

The IR sensors have been mounted and look absolutely wonderful. If you have been following the blog, you will have also seen the fingers that Shen designed that have been mounted so nicely on to the arm. Let's have some applause for Shen!

Knowing in advance how big the finger lengths were going to be was quite useful as I was able to limit down the sizes of the VeroBoard. We discussed different ways to mount the boards, but in the end the easiest solution that we still have right now is hot-glue; the interfaces between the curved surfaces were hard to adhere to as the angle didn't work very well. In the end, it's still quite sturdy and has survived the torture of our tests.

The picture above shows the (somewhat ugly) wiring of the sensors on the hand. As you can see, the ribbon cables are slightly raised up to then split off into Vcc, Ground and Analog wires.

PROBLEM: The raised ribbon wires were causing problems as they curled up and twisted around the hand, interfering with the movement of hte fingers and the wires of the other sensors.

SOLUTION: The Gauntlet - Aluminium plates that would guide the ribbon wires along the fingers. We did this for both the 'bones' of the finger. Funnily enough, the workshop had a stack of aluminium plates that were nearly the size we required that only necessitated small cuts using a guillotine. Where were they from? The metal backing from breadboards that the First-Years were given. What goes around comes around, hey?

Software

Just a final note on the IR sensors that we are using: how it responds to certain different types of material.

Here is a short table on the type of tests we did with the hand and the IR sensors:

Object	Thumb	Index	Middle	Ring	Pinky
All Fingers open	9	5	8	10	5
All Fingers closed	13	8	10	16	7
Red Ball	12	73	71	81	41
Roll of Masking Tape	35	16	21	30	7
White Foam	13	40	44	69	36
Pink Foam	13	63	73	92	52
Yellow Foam Mat	60	49	54	76	44

In most of these tests, it was only the index to pinky fingers that were holding the object - as you can see, the Thumb was not always involved in holding the object. Thus it has lower readings.

We also tried a black piece of foam, but, as with all our previous tests, it did not have a good response at all - it was pretty much the same as All Fingers Closed. Thus we did not include it in our tests - and it will not be part of demos. It is a limitation of the hand and the sensors.

Next post - a quick stroll alonger the shores of software for Vibration Motors