What would YOU do, given a 2-arm and lazy susan robot arm set?
Design News just covered an interesting idea...
... a tandem robot arm system by ST Robotics, that simply
combines a pair of arms on a turntable: (Very cool...)
Video: ST Robotics' 11-Axis Tandem Robot System
...
http://www.designnews.com/document.asp?doc_id=253426
Direct link to the embedded YouTube video:
...
http://www.youtube.com/watch?v=UXmYz2TOS6Q
It wouldn't be hard to do this, with a pair of Scorbot or other
cheap classroom training robot arms, and a simple rotary axis.
Taa Daa.. a similar robot, with MASSIVE reach!
(In fact, I already have a pair of Scorbot-IIIs and a linear axis.
I'm now seriously considering playing with this configuration
for my FMS, or at least keeping this idea "in my back pocket"... ;-).
Other benefits and possibilities, not mentioned.
1) There is inherent "part and tool swapping" capacity
in this configuration, WITHOUT the need for a complicated
(and heavy) end effector on a single arm.
Simply have one bot fetch the item or tool from the
station with a single gripper, then do a 180 with the
turntable and let the OTHER arm put the new part
or tool into place.
2) Next, having two lighter, MUCH simpler grippers in (1)
vastly increases the payload capacity of EACH arm,
allowing for the use of CHEAPER arms (vs loading an
arm down with a heavy, complicated "swapping" end
effector), while still retaining the "fast swap" efficiency.
3) For control, you simply cross connect I/O switches between
two basic single arm controllers to sync their programs
with each other. One does a move, then triggers a
subroutine on the other controller via the I/O bits.
The second arm then triggers the next move on the
first arm, etc. Or, a simple external micro can drive
BOTH single arm controllers to execute subroutines
as desired, via I/O handshaking.
IOW, they alternate moves or wait on each other,
using just the STANDARD arm controller boxes.
Yes, it will be up to you to watch for collisions
in your programming, but this does make it workable
with off the shelf controls, without requiring you to
create a custom, 11+ axis "dual-controller".
... VARIATIONS...
4) "Traveling Dual Arm" - Given an arm on a linear slide,
it wouldn't be difficult to simply add the turntable element
to the slide. You'd simply need some stabilizing casters
on both sides of the slide on the lower part of the turntable,
for tilt stability. This gives you not only far reach, but far
TRAVEL as well. IMO it would GREAT configuration for
servicing TWO parallel rows of CNC machines with one
dual-arm (or even SINGLE arm, see next item).
5) "Long Reach SINGLE Arm" - In fact, another thought is
to simply replace ONE of the arms on that turntable with
a simple counter weight. It would add a lot of "cheap reach"
to any SINGLE robot arm configuration.
... OTHER THOUGHTS ...
6) Parts can be handed off between the arms.
This may reduce the need for a large rotation angle on the
turntable, greatly simplifying the wiring around a rotary joint.
(eg only provide a 180+ degree table, vs a 360+ degree unit.
That is a MUCH simpler harness problem!)
7) Directly over the turntable's motor pivot position, you can
place a simple "part holding table", tool holder socket, or cup.
It is relatively dead space, and is a VERY handy, fixed,
known place in space for the robot arms to drop items
for exchange between them. It is also always in the SAME
relative position for each arm, regardless of table rotation,
or even linear axis travel below the turntable.
What do you all think of this arm system?
What would YOU envision doing, with such a system?
- Keith Mc.