COMMENT ⊗   VALID 00007 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	people and projects
C00003 00003	joint effort
C00004 00004	proposal
C00005 00005	Equipment
C00010 00006	SUMMARY
C00014 00007	BACKGROUND
C00021 ENDMK
C⊗;
people and projects

Sam Fuller; nice guy with power
Dieter Huttenberger; external research coordinator
Manufacturing Automation  Robotics
  Tom Williams, leader
  Ken McNaughton
  Leo Pinard
  Frank Glaser [student UMASS]
  3 full time people; 9 part time students UMass, etc
Inspection
  Dave Widder
  Darrell Briggs [Bedford]

couple of robots in production
  what kind?
  doing what?

  
They have the UMass vision system
  topics:   gray scale, range finding
  using zero crossings for edges
  scanning electron microscope images of lsi
    prototype device diagnostic and characterization
joint effort

investigate with IE modeling and simulation of the
  production system with DEC

task language for assembly; see below

proposal
  to Tom Williams
  
outline
  introduction
  research program
  equipment proposal
    equipment only, not field service
    try out a VAX 750 for dedicated vision.
    useful to have two Unibusses
    RM series disks faster than RP series
    LXY printer, dot matrix, for images
    DR11W in place of D11B; Grinnell supports it; small changes to board
    DEC adds on a few percent for extras overlooked
  payment schedule
  specific joint effort

later on, possible small support for student, very small
  e.g. one quarter or so; a year down the way
Equipment
Calculate service contract cost (no discount on this)

Disk drives
  *** include a disk with removable medium
  *** compare transfer rate of RM series with RP07; which is faster?
  DRIVE		TRANSFER   ACCESS/SEEK    CAPACITY   REMOVABLE   PRICE
                RATE	    TIME	    
  RM03-AA	1.2MB/s	   33.3ms          67MB      yes 	 $30,300/$20,300-12%
  RM80-AA	1.2MB/s    33.3ms         124MB      no 	 $29,900/$19,900-12%
  RM05-AA	1.2MB/s    38.3/30.3ms    256MB      yes	 $44,000/$34,000-12%
  RP07		1.3MB/s	   31.3/23.0ms    516MB      no	         $48,000/$38,000-12%
		2.2MB/s	

NEW high performance VAX
  after 2080; 2 1/2 years off


MEMORY
*** 64k memory chips available yet?  don't wait.
DEC
  quote $5k per megabyte; no discount on this; their catalog prices
    are as follows
    1 MB ECC memory MS780-DC        $13,800-12%
    2 MB ECC memory MS780-DD        $21,600-12%
    3 MB ECC memory MS780-DE        $31,100-12%
    4 MB ECC memory MS780-DF        $39,900-12%
Mostek  287-5080.
  .5MB 8016A00 	ECC	1 card			$1150 incl.the discount.
National; no 64k rams yet; 5MB 11/780; 1 MB 11/750; Donna Berry
  contact Charlie Wise 415-486-4504;
    had 2 boards for 1 month; like them and will buy more
  .5MB for 780	ECC; 1 card	225 nsec	$1100 Berkeley special price
  .25MB for 750					$845 Berkeley special price

UNIBUS
*** include a second UNIBUS
  DW780-AA $12,900-12%

Terminal and modem interfaces
  DEC DZ11-A	8 lines 	1 hex slot	9600 baud		$2570
  Able DH/DM	16 lines	1 hex slot	19,200 baud		$3700
  is it software compatible?
  do we need anything special for dialup lines?


VAX11/750 system
  SV-BXWAA-AA								$120,000
    RM80
    TS11
    DZ11-A    8 lines multiplexer
    1 MB memory
  1 MB additional memory						$5000
  floating point accelerator
  DR11-W								$1650


VAX 11/780 system							
  SV-AXVCB-CK								$319,100
    VAX 11/780
    3 MB memory
    2 memory controllers
    RP07 516 MB fixed media disk with controller with 2.2 MB kit
    TU78 tape drive with controller
    DZ11-A 8-line synchronous multiplexer
    LA 120 console terminal
    VMS license to copy
    H9602 cabinet for second memory controller
  FP780-AA floating point processor					10,600
  RP07-AA 516 MB fixed disk drive					38,000
  RP07-D 1.3 MB/sec to 2.2 MB/sec upgrade kit				N/C
  REM03-AA								$30,300
  DW780-AA second UNIBUS						$12,900
  DR11-W							2 at 	$1650
  insurance
  documentation
  DZ-11B 8-line multiplexer						$2050
  5 MB memory at $5000 per MB						$25,000 (no discount)
SUMMARY

The Robotics Laboratory of Stanford University proposes a joint effort with
Digital Equipment Corporation in automation for manufacturing.  In support of
the joint effort and the research program of the Robotics Laboratory, we request
a gift of 50% credit on a VAX computer system described in detail in a later 
section.  We also propose that DEC provide 50% credit on a VAX 11/750 system
to be used as a dedicated processor for vision and inspection in conjunction
with manipulation.  We further propose that DEC provide 50% credit for a
second VAX 11/750 system to be used for a dedicated robotic workstation 
controller.

PROPOSED EFFORT

We propose to study language systems for workstation controllers
in integrated  manufacturing systems.
We plan to study examples of how such language would be used in an
industrial environment, how it can be used by manufacturing personnel,
and special SCRIPT-like language for their use.
We plan to study the hardware and software interfacing of industrial workstations.
We plan to investigate synchronization, debugging, and world modeling 
in languages for workstation controllers.

Some of our ongoing research efforts may be of interest to DEC.  We propose to:
[1] provide DEC with direct access to results of our research;
[2] to consider DEC's problems in our research on automation, and to test
and evaluate some of our systems on examples typical of DEC's problems; and
[3] to assist DEC in limited ways to transfer some of our results to their
facilities.

APPROACH

We propose to study these issues in relation to extensions to the AL system
for programmable automation.  Stanford has developed AL, and has made a portable
version in PASCAL on a PDP11/45 running RSX-11.  We plan to see to what extent 
modifications and extensions of AL are natural in order to function as a workstation
controller.  We will compare AL with MCL and AML, and with ADA for its synchronization
capabilities.

We are not aware of an adequate language for describing mechanical tasks and assembly
operations.  
We have implemented MSM as a system based on MTM, a widely-used system for analysis
of human tasks and assemblies.  MSM embodies ideas and implementation toward
a system which formalizes assembly and which is natural to manufacturing
engineers.  MSM is implemented in AL.  
We have also carried out kinematic analysis of some assembly operations.

BACKGROUND
RESEARCH PROGRAM OF THE ROBOTICS LABORATORY

The Robotics Laboratory carries out a broad research program in computer vision
and manipulation, including:
    design: design of a three finger hand; design of advanced manipulators;
    dynamics and control of manipulators: multiple, cooperating manipulators;
      end point sensing and visual tracking; adaptive control; 
    task level force sensory control: sensor technology; interfacing;
      analysis of force strategies; high level language for force control;
      teaching force strategies;
    programming systems for programmable assembly: the AL system has an interpreter,
        force control, source language debugging; graphics for debugging and experimentation
      with force control; 
    next generation of programming systems for robots:  geometric modeling 
      with geometric reasoning;
      offline programming; integrating sensing and vision;
    inspection: development of an intermediate level industrial vision system with
      less sensitivity to lighting, greater discrimination of shape, however 2D;
    ACRONYM integrated vision system:  ACRONYM combines geometric modeling with
      geometric reasoning in the form of a rule-based system; ACRONYM integrates
      interpretation with stereo vision and ranging, and includes edge-based segmentation.
    VLSI implementation of vision operations

History

The robotics program at Stanford was begun in 1965 by John McCarthy at the
Stanford Artificial Intelligence Laboratory.  

Manipulator design:  Currently we are collaborating with 
Prof Bernard Roth, Mechanical Engineering Dept, and with Carl Ruoff, JPL,
on design and control of a three-finger hand with nine degrees of freedom,
a "universal gripper".  The
motivation for the hand is to provide fine control of forces in assembly
and tool using, and to provide the relevant degrees of freedom for repositioning
objects after grasping for subsequent assembly operations.  This is a research effort;
we expect that further effort will be necessary to make a 
rugged and reliable hand that can be used in everyday work.
Conceptual design including theoretical work, and detailed design are complete.
The fabrication of the hand is just beginning.
Work in optimizing the design of the hand has resulted in 
significant theoretical contributions [Salisbury 81] and form the basis of
the forthcoming PhD thesis of K. Salisbury.

Sensory Technology:  Historically, we have done considerable work in 
technology for force and contact sensing [Binford 72]. Our 
Scheinman force sensing wrist [Scheinman 74] is one of the first
successful wrists to measure force-torque to high accuracy.
The wrist has been a basic component in industrial automation studies
involving two manipulators working in synchrony [Shimano, 
sprinkler].
We are about to begin new work in collaboraton with Prof Larry Leifer of
Mechanical Engineering, Prof Robert Cannon of Aeronautics and Astronautics,
and Prof John Shaw of Applied Physics on contact sensors using piezoelectric
polymers and array techniques.  We also plan to make integrated circuit
versions of sensor interfaces for strain gage sensors for the Scheinman
force sensing wrist, using capabilities of Stanford's Integrated Circuits lab,
Department of Electric Engineering.  We are currently designing force-sensing
fingers to complement force sensing wrists, to provide greater sensitivity
to small forces, and to minimize the mass between sensor and load to minimize
inertial forces caused by accelerations.\.

\JPUMA consortium:  We have organized a consortium of users of the Unimate PUMA
to instrument the PUMA robots.  Stanford is taking the lead in providing 
resources for a pooled effort to provide: electric hands with two fingers;
wrist force sensors; and finger sensors.  A hand is about to be fabricated
at Stanford. 
Detailed
plans and schedules are available on request.  The effort will have the benefit
of providing common equipment and pooling effort while saving cost.
The hand to be used is not the same as the three-finger hand which was described
above.  For the consortium, a conservative design with two degrees of freedom
is being made.  It is designed to be modified to a three finger version with three 
degrees of freedom.\.