perm filename PRNBS1.V[1,VDS] blob
sn#104634 filedate 1975-02-26 generic text, type C, neo UTF8
COMMENT ā VALID 00003 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00002 00002 TECHNICAL PROPOSAL
C00013 00003 PRICES
C00014 ENDMK
Cā;
TECHNICAL PROPOSAL
To meet the specifications of solicitation No.4-35921, VICARM
proposes to deliver a modified version of our standard model
"Stanford" six degree of freedom computer controlled manipulator. A
specification sheet for this standard manipulator is enclosed as
Attachment 1.
The paragraphs which follow detail the general properties of
our model Stanford manipulator and describe the proposed
modifications to enable it to meet the requirements of the
solicitation. Attachments A, B, and C are layouts of our model
Stanford manipulator with some of the proposed changes sketched in.
1) The standard manipulator has the required geometry and is
an option A type, having a prismatic third joint. Base up (inverted)
installation is allowable for the standard manipulator. To completely
meet the requirements of section 3.1, the wiring paths of the arm
will have to be redesigned to allow for internal routing of the seven
instrumentation wires. This will require extra shrouding to allow
the wires to pass around joint 5, a special housing for shrouding
the wire loop around the prismatic joint and also a redesigned path
thru joints 2 and 1. The standard model wire paths are shown in
Attachments A and B, along with a possible redesigned layout. As
redesign to provide entirely internal wire paths will entail quite a
bit of work, the details of this have been left out of this proposal.
It is anticipated that this will be a negotiable point, as no mention
is made of a requirement that motor, brake, potentiometer, or
tachometer wires be internal. The limits of motion, maximum speed,
and geometry of our standard arm are such that these particular
specifications are met with no redesign, provided the duty cycles
are no greater than typified by the proposed acceptance tests, and
minimum velocity specifications are for an unloaded arm. To meet the
minumum velocity specifications at full load will require substantial
redesign.
2) Our proposed terminal device is our standard parallel jaw
hand (see Attachment C), which meets the requirements of section 3.2.
This hand has an integral motor,gearbox, brake and potentiometer.
Jaws are guided on steel rods and driven by a single central pinion
and double racks.
3) To meet the requirements of section 3.3, which specifies
an accuracy and precision within a working volume much larger than
the defined workspace of our standard manipulator(see specs. on
attachment 1), the arm will have to be made with closer machining
tolerances. The potentiometer elements on joints 1,2 and 3 will also
have to be changed to improve their linearity,and noise level to meet
the increased resolution and accuracy requirements. Joints 4,5,6 and
the hand will keep our standard integral potentiometers.
4) We strongly suggest that the use of optical encoders be
considered instead of potentiometers on joints 1,2, and 3 as these
devices are inherently highly linear, can have more than 12 bit
resolution, and interface directly to the i/o buss rather than thru
an a/d. Our standard manipulator is available with 13 bit incremental
encoders on the first 3 joints with a resulting increase in
resolution and accuracy as noted in our specification sheet. We have
suggested but not proposed optical encoders only because the
solicitation requires that the sensor interface to an a/d and it is
not practical nor realistic to consider using a d/a to get encoder
output into a form suitable for an a/d. A third alternative is the
use of resolvers as joint position sensors.
5) To satify another requirement of section 3.3 that the
manipulator follow an arbitrary straight line trajectory at any reach
and load up to the maximums specified, will require the addition of
forced air (fan) cooling of the motors. This is because conditions
of maximum torque and large duty cycle can be encountered in this
mode; as in the case of a very long and slow motion at maximum load.
6) To insure meeting the task performance, overshoot, and
stiffness specifications, the servo system will be a combination
hardware and software servo system. All 6 joints will have both
analog tachometer and position feedback. At the present time we
suggest implementing high frequency velocity feedback in hardware,
but position and low frequency velocity feedback thru the computer. A
completely hardware servo system is also proposed to simplify
installation, checkout and the acceptance testing of the
manipulator, as this will easily interface with the existing NBS
teach mode software.
7) Software for control of a manipulator of this type
presently exsits at the Stanford University Artificial Intelligence
Laboratory (on the ARPANET). Current software is for use with a
PDP-10 and PDP-6. Both of these computers are 36 bit word, floating
point processors. Servo code is presently being written for use on a
PDP-11/45, and is expected to be in useable form before the end of
this year.
8) Per section 3.5, joint 6 will be redesigned to give it a
maximum torque equal to that of joint 5, while maintaining a gear
ratio which will still allow it to meet the speed requirement of 8
radians/sec. This will require a larger motor, stronger gearing, and
a larger housing. It is observed that the peak power requirements
represented by the specifications imply that the motor of joint 6
should have twice the power of joint 5. As an alternative, it is
possible to increase the strength of the brake in the standard joint
6. This could allow the braked torque of the standard joint 6 to
equal the maximum torque of joint 5.
9) Manual Control. The manipulator will be an "Option C"
type with brakes on all six joints,and the hand. All joints will
have either harmonic drive or spur gear gearboxes which will allow
them to be backdriven manually with the brakes off.
Several specifications presented in the RFP have not been
discussed above. This is because our standard arm meets these other
requirements with little or no modification. In any event it may be
taken that provisions have been made for safely accomodating
overloads or full scale command errors and meeting the interface
connection requirements, etc.
PRICES
Complete manipulator including ALL the requirements of the
solicitation. - $37,500
Manipulator as above execpt without- Internal wiring specs(paragraph 1)
Min. velocity specs at full load.
$27,500
Manipulator as in B above, execpt without
- high strength joint 6, but with high strength
brake(paragraph 5).
$24,500
These prices include the acceptance tests at NBS facilities per section
5.0 of the solicitation, all the
specified documentation in section 4.0., and delivery within 150 days
of award of contract.