What is the picture of?

[bk2]

Hi,
I would like to know how one can get hold of a robot arm like the one pictured in the third last drawing in the Robot Arm Tutorial at URL http://www.societyofrobots.com/robot_arm_tutorial.shtml .

It is the picture right above the cartoon "Die Man".

[Metal Slug 2]

Since that picture is under the "Haptic sensing" header, and that a human seems to be manipulating the device, I would assume that that picture is not actually a robotic arm, but rather a model arm used to control a larger robotic arm. from the page: "a miniature model with position sensors"

[Gertlex]

For others' reference in helping, he's referring to this picture/arm:

[bk2]

Thanks for including the picture. The reason for asking is that I am looking for a simple robot arm - without any sensor or hand - that would be able to move from side to side in a "Union Jack" pattern (up - down, side to side, and the two diagonals) and a few other linear movements in other planes. The end of the arm is to acts as a focal point.

Does anybody know of such one?

[Soeren]

Hi,

I am looking for a simple robot arm - without any sensor or hand - that would be able to move from side to side in a "Union Jack" pattern (up - down, side to side, and the two diagonals) and a few other linear movements in other planes. The end of the arm is to acts as a focal point.

Wouldn't an X/Y-table do for that?

[bk2]

Sorry, I do not know what an x/y-table is, so I would not know.

The application of it is such that a person is sitting in front of the arm - looking at its end, while it is moving. The movement is either in a plane in front of the person (like if you had a computer screen in front of you, but since we need to a range of +/-50 degrees right left and 20 up and 30 degrees down at a 40 cm distance, it would have to be rather large - thus too expensive expensive).

Also, we need the arm to move to and from 50 cm to 2 cm distance along a line from between the person's eyes and outwoards approx 0 - 10 degrees down. The apparatus (arm or whatever) should not bee prohibitatively expensive to produce, and should not block cameras standing behind the arm, acting as eye trackers.

[Gertlex]

Hire an engineer to do the job right

[bk2]

Would do - if we were to start production planning. But we will be doing research for a while - and research councils, do not pay for engineers in this case  :-(

So we wanted to see if there is a suitable robotarm available, that we can use in this phase.

[Soeren]

Hi,

The application of it is such that a person is sitting in front of the arm - looking at its end, while it is moving. The movement is either in a plane in front of the person (like if you had a computer screen in front of you, but since we need to a range of +/-50 degrees right left and 20 up and 30 degrees down at a 40 cm distance, it would have to be rather large - thus too expensive expensive).

Also, we need the arm to move to and from 50 cm to 2 cm distance along a line from between the person's eyes and outwoards approx 0 - 10 degrees down. The apparatus (arm or whatever) should not bee prohibitatively expensive to produce, and should not block cameras standing behind the arm, acting as eye trackers.

Just to be 100% certain...
The point to focus on will only have to move like if it was on a large screen (i.e.left/right and up/down)?
Or is there 2..50cm in the Z-axis (towards/away from the test subject)?

If you make a narrow "board" with a pulley wheel in each end and use a small stepper motor to turn one of them (with a wire loop or similar running between the pulleys). you have one axis of movement (like what's moving a printer head or a flatbed scanner head), call this the side to side (X) axis. Mount on this another board that is pretty much the same (just change the length of the board and that will have movement in the up-down (Y) axis. This can be mounted at the edge of table top and the pointer/spot can be placed on a thin piece of spring steel rod of say 2.0mm to 2.5mm, so won't block the view like most other methods.
If you need a Z-axis movement as well, this is probably best made by tilting the steel rod back and forth with a slightly different setup, but shouldn't be hard at all.

Hope it didn't sound too confusing 

[bk2]

Not at all - if I understood you right :-) You did put my thoughts in a totally different direction, which was probably a very good idea. Less complicated and faster to build. Thanks a lot.

The idea of having the focal point travelling on a thin but rigid "board" controlled by a step motor is excellent. The board can then be monted according what experiment one wants to do - up/down, diagonally, left/right - even forth and back. Probably only one motor is needed for all those, since we have only planned linear movement tests so far.