 |
Description  |
|
|
FIELD OF THE INVENTION
The present invention relates to tactile sensors, and more particularly,
the present invention relates to tactile sensors which are particularly
suited for use with automated equipment of various types, including robots
and robot end effectors.
BACKGROUND OF THE INVENTION
Increasingly sophisticated types of automated equipment, such as robots,
are being used in a wide variety of manufacturing operations, including
inspection, identifiction, pick and place procedures, assembly procedures,
and the like. To enable robotic equipment to grip parts in a particular
manner without damaging the same, some robots are being equipped with end
effectors, or jaws, having tactile sensors which give the end effectors a
certain degree of feelability. Other automated equipment utilizes work
surface mounted sensors to provide sensory feedback to assist in the
manipulation of parts thereon. The tactile sensors are electrically
connected to suitable circuitry associated with the equipment to enable
various parameters of a part, including its size and shape, and the force
exerted on the part, to be detected.
BRIEF DESCRIPTION OF THE PRIOR ART
Tactile sensors which operate on various principles are known. For
instance, a satisfactory optoelectric tactile sensor manufactured by Lord
Corporation of Erie, Pa. has a flexible platen with an array of depending
projections which cooperate with light emitters and receivers disposed on
opposite sides of the projections to provide electrical outputs when the
portion of the platen superadjacent the projection is deflected and light
transmission between the emitters and receivers is modulated by movement
of the projections.
A tactile sensor which utilizes a pressure responsive electrically
conductive elastomer is disclosed in U.S. Pat. No. 4,014,217 to Lagasse et
al. This sensor has a series of annular electrodes which surround a series
of central electrodes and which cooperate with a doped elastomeric member
to produce a readout when current fields, established in the elastomeric
member by the electrodes, are altered by applied pressure.
One form of tactile sensor which incorporates discrete semi-conductive
sensing sites is disclosed in U.S. Pat. No. 4,492,949 to Peterson. This
sensor includes a compressible panel having a series of pressure
responsive electrically conductive posts distributed in an array between a
series of rows of partially exposed conductors embedded in a flexible
member overlying the panel and a series of columns of partially exposed
conductors embedded in a base underlying the panel orthogonal to the rows.
The posts are located at the intersections of the rows and columns so that
when pressure is applied superadjacent each post, its electrical
resistance decreases, permitting current flow from a row to a column. The
rows and columns are sequentially scanned by suitable cross-multiplexing
circuitry to provide a readout of the location and magnitude of applied
pressure.
The utilization of cross-multiplexing electronic circuitry in combination
with tactile sensors, such as described in the above-referenced patent to
Peterson, the disclosure of which is incorporated by reference herein,
provides the advantage of enabling the density of sensing sites to be
increased and thereby to increase the resolution of the sensor. A limiting
factor in achieving high resolution, however, has been the introduction of
electrical cross-talk, or phantom switching phenomenon, in cross
multiplexed sensor systems. To overcome this problem, some sensors have
diodes located at each sensing site, such as disclosed in U.S. Pat. No.
4,481,815 to Overton.
For a more complete review of the state of the art with respect to tactile
sensing using conductive elastomers, reference is made to the following
articles: Peter and Cholakis, "Tactile Sensing For End-Effectors", Barry
Wright Corporation, Watertown, Mass.; Marc H. Raibert and John E. Tanner,
"Design And Implementation Of A VLSI Tactile Sensing Computer", The
International Journal of Robotics Research, Volume 1, No. 3, Fall 1982;
William Daniel Hillis, "Active Touch Sensing", Massachusetts Institute of
Technology Artificial Intelligence Laboratory, April 1981, AI Memo 629,
pp. 1-37; and Purbrick, John A., "A Force Transducer Employing Conductive
Silicone Rubber", Proc. 1st, Robot Vision and Sensors Conf., IFS Pubs.,
Ltd., Kempston, Bedford, England, 1980, pp. 73-80.
Prior art tactile sensors have various drawbacks. Those utilizing
optoelectric principles are expensive to manufacture. Those utilizing
electrically conductive elastomers have low sensitivity and a proclivity
to wear because the conductive particles doping the elastomer affect
adversely its mechanical properties. Those sensors utilizing discrete
pressure sensitive posts in a cross-multiplexing array, and those
incorporating diodes, are expensive to manufacture. Furthermore, each of
the prior art sensors has limited resolution and sensitivity, and none is
as durable and inexpensive to manufacture as desired.
OBJECTS OF THE INVENTION
With the foregoing in mind, a primary object of the present invention is to
provide a novel tactile sensor which overcomes the limitations of known
tactile sensors.
Another object of the present invention is to provide an improved tactile
sensor which provides both high resolution and sensitivity in a large
field.
A further object of the present invention is to provide a rugged tactile
sensor which is relatively inexpensive to manufacture.
As a still further object, the present invention provides a method and
apparatus for accurately sensing pressure distribution over a relatively
large area with a minimum of components.
As yet another object, the present invention provides a tactile sensor
which operates in a cross-multiplexing mode with a minimum of interference
caused by electrical cross-talk and phantom switching.
SUMMARY OF THE INVENTION
More specifically, the present invention provides tactile sensing apparatus
which functions in a novel manner to provide an accurate readout of both
the magnitude and distribution of pressure applied to a sensing surface.
The sensor comprises a resilient platen having an upper surface adapted to
be contacted by an object and a lower surface overlying a plurality of
spaced pressure sensing sites. Each site includes emitter electrode means
and companion collector electrode means surrounding the emitter electrode
means in spaced relation therewith. Flexible conductive means on the lower
surface of the platen cooperates with the emitter and collector means to
cause substantially all of the current emitted by an emitter means to be
collected by its companion collector means when the platen is deflected
downwardly superadjacent the sensing site and a voltage differential is
applied across the emitter and collector means.
Preferably, the sensing sites are provided by a series of current emitter
electrodes electrically interconnected in a row by a bus bar and a series
of current collector electrodes surrounding the current emitter electrodes
and electrically connected in a column by another bus bar disposed above
and across the row. The emitter electrodes are electrically insulated from
their companion collector electrodes, and the bus bars defining the rows
and columns are electrically isolated from one another. The rows are
provided on the underside of an insulated panel, and the columns are
provided on the topside thereof coplanar with the emitter electrodes. A
resilient platen overlies the array of emitter and collector electrodes
and has on its underside a thin, rough coating of electrically conductive
material engaging the emitter and collector electrodes. A means is
provided for simultaneously supplying a positive voltage to a selected one
of the rows of emitter electrodes and a lower voltage to the columns for
causing current to flow at a site under pressure from one of the emitters
at such site, through the coating, and to the companion collector for
collection and measurement by multiplexing circuitry which repeats these
steps periodically to produce a readout of both the location and magnitude
of applied pressure.
BRIEF DESCRIPTION OF THE DRAWINGS
The foregoing and other objects, features and advantages of the present
invention should become apparent from the following description when taken
in conjunction with the accompanying drawings, in which:
FIG. 1 is a fragmentary perspective view illustrating a few different
applications for a tactile sensor embodying the present invention;
FIG. 2 is a plan view of a tactile sensor embodying the present invention
with a portion of its sensing surface peeled back to reveal certain
interior construction details;
FIG. 3 is a greatly enlarged fragmentary plan view of the revealed portion
of the sensor illustrated in FIG. 2;
FIG. 3A is a schematic representation of some of the FIG. 3 sensor circuit
components in association with control components and circuitry;
FIG. 4 is an enlarged fragmentary cross-sectional view taken on line 4--4
of FIG. 2;
FIG. 5 is a greatly enlarged sectional view of the area contained within
the circle illustrated in FIG. 4;
FIG. 6 is an enlarged somewhat schematic fragmentary plan view of certain
sensor elements;
FIG. 7 is a sectional view taken on line 7--7 of FIG. 6;
FIG. 8 is an electrical schematic diagram illustrating certain principles
of operation of the present invention;
FIG. 9 is an enlarged fragmentary plan view of a portion of a modified
embodiment of the present invention; and
FIG. 10 is a plan view of another modified embodiment of the present
invention.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
Referring now to the drawings, FIG. 1 illustrates a tactile sensor 10 which
embodies the present invention. The tactile sensor 10 may be supported on
a work surface 11 for sensing the orientation of a part P in an assembly
operation; or, a smaller version 10' thereof may be mounted on one or both
movable elements 12 of an end effector 13 mounted at the end of a robot
arm 14 to enable the part P to be gripped in the proper manner with the
proper amount of pressure.
As discussed heretofore, known tactile sensors are limited in resolution,
sensitivity, durability and manufacturability. In addition, known tactile
sensors utilizing cross-multiplexing techniques to maximize field size and
minimize components have been plagued by electrical cross-talk and phantom
switching problems which may be described briefly as the proclivity for
electrical activity at one pressure sensing site to produce an output at
an adjacent pressure sensing site. Attempts to overcome these problems
either by complex mechanical structures or by connecting electrical
components in the structure have not been entirely satisfactory.
The tactile sensor of the present invention overcomes the limitations of
known prior art tactile sensors. For instance, it has high spatial
resolution and fine sensitivity over a relatively large sensing area. In
addition, it is durable and capable of being manufactured readily.
Furthermore, tactile sensing apparatus embodying the present invention
functions accurately to sense objects by eliminating electrical cross-talk
among its various sensing sites.
Referring again to the drawings, and particularly to FIG. 2 thereof, the
tactile sensor 10 of the present invention comprises a frame 15 which may
be of any shape but which, in the illustrated embodiment, is square. A
resilient platen 16 is mounted to the frame 15 by a layer of adhesive
extending around its peripheral margin. The platen 16 has an obverse
surface, or topside 16a, adapted to engage an object, such as the part P
illustrated in FIG. 1, and a reverse surface, or underside 16b. One or
more wiring harnesses, such as the wiring harness 17, are povided for
connecting the sensor 10 to suitable circuitry in a manner to be
described.
The tactile sensor 10 has high resolution. To this end, an array of closely
spaced pressure sensing sites, such as the sites S.sub.1, S.sub.2 and
S.sub.3 (indicated within the perimeter of the circle denominated "FIG. 3"
in FIG. 2) are provided within the frame 15 underneath the platen 16. The
field, or area, occupied by the sensing sites S.sub.1 -S.sub.3 is
substantially square and coextensive with the area of the platen 16.
Preferably, the sensing sites S.sub.1 -S.sub.3 are located at equally
spaced horizontal and vertical intervals having, by way of example, center
to center spacings on the order of less than about 0.100 inches, and more
preferably, about 0.080 inches, thereby enabling 6400 sensing sites to be
arranged in a field of less than 40 square inches for a sensing site
density in excess of 150 sites per square inch.
According to the present invention, pressure sensing is accomplished by
powering the emitter electrodes at the sites and measuring the current
collected by their respective companion collector electrodes. As best seen
in FIG. 6, each sensing site, such as the central site S.sub.2 comprises
an emitter electrode 20 surrounded by a companion collector electrode 21
disposed coplanar therewith and spaced therefrom by a continuous annular
gap 22 of a very narrow width, such as 0.008 inches. A continuous,
flexible conductive coating, or layer, 23 (FIG. 5) is disposed on
substantially the entire undersurface 16b of the platen 16, except at its
peripheral margin 16c, to provide omnidirectional resistivity on the
platen undersurface 16c and to cooperate with the the platen 16 to provide
a plurality of microprotrusions forming an asperity, or roughness, on its
undersurface 16b. Thus, when the platen 16 is deflected downwardly into
engagement with electrodes 20, 21 at an underlying site, such as the site
S.sub.2, the layer 23 bridges the gap 22 between the electrodes and
conducts current from emitter electrode 20 to collector electrode 21.
The magnitude of the current conducted from emitter electrode 20 to
collector electrode 21 is a function of both fixed and variable
parameters. The fixed parameters include: (1) the lateral resistance of
layer 23, i.e. along the platen 16, which is relatively great when (as in
the preferred illustrative embodiment) the layer is very thin; (2) the
relative sizes in a lateral direction of the narrow gap 22 and the
relatively wide collector electrode 21; and (3) the lateral dimensions of
the gaps or spaces 25a, 25b between the collector electrode and any
adjacent collector electrode, such as the gap 25a between the central
collector electrode 21 and the collector electrode 21a of the right hand
sensitive site illustrated in FIG. 7. A variable parameter affecting the
magnitude of the current conducted from emitter electrode 20 to collector
electrode 21 is due to the microprotrusions that impart an asperity, or
roughness, to the layer 23.
When the pressure effecting engagement between layer 23 and the underlying
pair of emitter and collector electrodes is small, the microprotrusions
provide only a small area of contact between the layer and the electrodes.
The resistance to current flow from emitter electrode 20 to collector
electrode 21 is therefore high, and the current flow is low. A greater
pressure upon platen 16 increases the area of contact of the layer with
the electrodes 20, 21, thus decreasing the resistance and increasing the
current flow from emitter 20 to collector 21. By monitoring the current
received by the collector 21 at a particular site, the pressure upon such
site can be ascertained and a grey scale, or analog, output signal
produced, provided that such electrode receives only current from its
companion emitter electrode 20 and not also, or instead, from an emitter
electrode at some other site.
To satisfy the foregoing condition, to prevent electrical cross-talk among
the various sensitive sites, and to maximize the available current signal
strength, substantially all of the current received by layer 23 from an
emitter electrode at a particular site is transmitted to and into the
companion collector electrode at such site. No significant amount of such
current is transmitted to any other site. This results in highly desirable
site localization of electrical activity of the emitted current which, in
part, is attributable to the emitter electrode at each site being totally
encircled or surrounded by its annular companion collector electrode. Site
localization of electrical activity is also attributable to the differing
resistances to current flow that are present within and between the
sensitive sites by reason of the component size and spacing parameters
previously described.
The differing resistances to current flow are schematically illustrated in
FIG. 8, to which reference is now made. Resistances R.sub.1, R.sub.3 and
R.sub.5 represent variable ones whose magnitude is a function of the
pressure imposed upon the platen 16. The magnitude of resistance R.sub.3
is also dependent upon the lateral, or radial, dimension of collector
electrode 21, as is the resistance R.sub.5. Resistance R.sub.2 has a
magnitude determined by the resistivity of layer 23, by the lateral
dimensions of the gap 22, and by the pressure applied to that portion of
platen 16 overlying electrodes 20, 21. The magnitude of resistance R.sub.4
is dependent upon the resistivity of layer 23, upon the size of the
lateral gap or spacing between adjacent collector electrodes, at any
adjacent sensitive site, such as the gap 25a between the collector
electrode 21 and the collector electrode 21a, and upon the pressure
applied to the overlying platen 16 at the sensitive site in question and
at the adjacent site or sites. The resistance R.sub.3 under all operating
conditions is much less than the sum of the resistances R.sub.4 and
R.sub.5, i.e. R.sub.3 <<R.sub.4 +R.sub.5.
Thus, when emitter electrode 20 is energized from a suitable power source
30, and a pressure (such as indicated in FIG. 7) is applied upon the
portion of the platen 16 overlying such electrode 20 and its companion
collector electrode 21, substantially all of the current conducted from
emitter 20 follows the path that includes the resistances R.sub.1, R.sub.2
and R.sub.3, extending to collector 21, rather than any alternative path
(such as the illustrated one which includes resistances R.sub.1, R.sub.2,
R.sub.4 and R.sub.5) extending to an adjacent sensitive site. In other
words, the resistance between the emitter 20 and an adjacent non-companion
collector 21a, and the resistance between the emitter 20a and an adjacent
non-companion collector 21, is greater than the resistance between either
emitter electrode 20 or 20a and its respective companion collector
electrode 21 or 21a. Amplifier 32 monitors current received by collector
21 and provides an output signal voltage representative of the magnitude
of the applied pressure. The amplifiers 31-33 also function in a
conventional manner to apply virtual ground electrical potential to their
respective collectors.
In order to enable the location of pressure application to be determined,
and to simplify the sensor design and support circuitry, the sensor 10 is
designed to be operated in conjunction with cross-multiplexing means, such
as the type of circuitry disclosed in the above referenced Peterson
patent, the disclosure of which is incorporated by reference herein. To
this end, the emitter electrodes, such as the electrodes, 20, 20a and 20b,
are each electrically interconnected to a bus bar 35 (FIGS. 3 and 3A),
provided on the underside of an insulated panel 38. In the present
instance, such connection is provided by transverse pins, such as the
hollow pin 39 (FIG. 5) projecting upwardly from the bus bar 35 through the
panel 38. A series of bus bars, such as the bars 36 and 37, extend in rows
in spaced parallel relation with the bar 35 in the manner illustrated in
FIGS. 3 and 3A. The bus bars 35-37 are electrically isolated from one
another along their lengths by small air gaps 40, 41 to define horizontal
rows of electrically interconnected emitter electrodes.
The collector electrodes, such as the electrodes 21a, 21 and 21b, are
electrically interconnected on the topside of the panel 38 by means of
conductive bus bars 42, 43 and 44 which form the electrodes, such as in
the manner the electrode 21 of sensing site S.sub.2 (FIG. 6) is formed
integral with, and thus connected to, the bus bar 43. The bus bars 42-44
are arranged in parallel relation in vertical columns and are separated
laterally from one another along their lengths by small air gaps 45, 46,
respectively providing electrical isolation therebetween. The collector
bus bars 42-44 are disposed at right angles to the emitter bus bars 35-37,
and the sensing sites S.sub.1 -S.sub.3 are located at the intersections
thereof.
The insulating air gaps 45 and 46 between the columnar bus bars 42-44, and
the corresponding air gaps 40 and 41 between the rows of bus bars 35-37
are small, each gap being on the order of 0.008 inches. Preferably, the
rows and columns of bus bars, such as the bus bars 42-44, are provided
with extensions 42a-44a which terminate in plated pin holes in the margin
of the panel 38 to permit the wiring harnesses 17 to be connected by
conventional connectors. The reverse surface of the platen 16 is left
uncoated around its peripheral margin at 16c to provide electrical
insulation above the bus bar extensions and pin connections. The emitters,
collectors and their respective bus bars are provided on the panel 38 by
conventional printed circuit manufacturing processes which involve plating
through holes in the panel 38 to provide the emitter electrode pins and
then etching away conductive material to provide the referenced air gaps
and to define the bus bars and the current emitter and collector
electrodes.
As is indicated in FIG. 3A the control circuitry and components utilized in
the present invention include cross-multiplexing means in the form of a
multiplexer 60 that applies a positive voltage to a selected one of the
bus bar rows, such as the bus bar 36, and a demultiplexer 62 that applies
a lower voltage, preferably at virtual ground potential, to each of the
bus bar columns, or to at least a three unit group such as that comprised
of the bus bars 42, 43 and 44. This establishes a pressure responsive
current flow path through the conductive layer 23 between each emitter
electrode of the selected bus bar and its companion collector electrode at
a site, such as at sensing site S.sub.2, under applied pressure. Current
flow in successively selected ones of the columnar bus bars, such as bar
43, is directed to an analog to digital converter 64, which may be an
array of operational amplifiers such as shown in FIG. 8, that converts the
current to a proportional voltage signal. The voltage signal is
transmitted from converter 64 by demultiplexer 62 to a computer 66 that
controls or at least monitors operation of multiplexers 62, 64. The
computer correlates each voltage signal and site location, and provides an
appropriate readout or display identifying the location or identity of
each site S along the active emitter bus bar row that is under an applied
pressure, and identifying the magnitude of such pressure. These steps are
repeated for each other row on a row by row basis.
While all of the columns may be held to virtual ground potential during
multiplexing, only three need be. For instance, only the pair of columns
42 and 44 alongside the column 43 being read must be held at ground
potential. Therefore, groups of three columns may be scanned across each
row when activated.
The emitter and collector electrodes may be arranged in a pattern such as
illustrated in the portion of a sensor 110 of FIG. 9 wherein a common
collector electrode, such as the electrode 121, surrounds a group of three
emitter electrodes, such as the emitter electrodes 120, 120a, and 120b.
With this arrangement, when voltage is applied to one of the rows, such as
the row 135 connected to the upper emitter electrode 120b, current
collected by the collector electrode 121 is representative of the pressure
applied to the platen 116 superadjacent the emitter electrode 120b. By
activating each row sequentially and measuring current flow from each
column while a row is activated, pressure applied at each sensing site may
be determined.
If desired, the rows and columns of emitter electrodes and collector
electrodes may be arranged in various other patterns than described
heretofore, such as the circular pattern illustrated in the modified
sensor 210 of FIG. 10 wherein the rows 235-237 extend radially and the
columns 242-244 extend circumferentially about an open center.
The tactile sensor of the present invention is capable of being
manufactured readily. To this end, the platen 16 is fabricated from a
sheet of relatively soft elastomeric material, such as natural or
synthetic rubber. The underside of the platen 16 is spray coated with an
electrically conductive polyurethane which comprises conductive particles
in an elastic carrier, such as H322 and L300 manufactured by the Chemical
Products Group of Lord Corporation, Erie, Pa. The spray should deposit the
polyurethane in as thin a coating as possible, and preferably the coating
should have a thickness of less than about one mil. The microprotrusions
provided by the coating should have a roughness which is less than that
which would be visible to the naked eye or palpable by touch.
The sensor 10 can be assembled readily. To this end, as best seen in FIG.
4, the panel 38 is mounted beneath the conductive coating 23 on the
underside of the platen 16 by a rigid insulating board 50 which underlies
the panel 38 and extends substantially coextensively with the area of the
platen 16. The insulating board 50 is supported on an inner frame 51
which, in turn, is fastened to a base 52 to which the frame 15 is secured
by means of a spacer 53 and fastener 54. This arrangement permits the
various electrical connections to be provided in the margin of the panel
38 at spaced peripheral locations, such as illustrated in FIG. 3.
In view of the foregoing, it should be apparent that the present invention
now provides an impr | | |
