Note: Descriptions are shown in the official language in which they were submitted.
WO91/10981 P~T/US91/004~
2 ~ 7 ~
Description
POSITION DETERMINING APPARATUS AND METHOD
FIELD OF THE INVE~TION
This invention relates to graphical data apparatus and,
more particularly, to an apparatus and method for determining
the position of a movable element in a data space.
BACKGROUND OF THE INVENTION
Graphical digitizers are conventionally used to input
graphical coordinate information, or the like, to a companion
system. In a graphical digitizer, wave energy is typically
passed between a movable element (such as a stylus or cursor)
and transducers located at fixed reference locations. The
transit time of the wave energy traveling (in either
direction) between the movable element and the reference
locations is used in determining the position of the movable
element, typically in terms of digital coordinates. A type
of graphical digitizer manufactured and sold by the assignee
hereof, Science Accessories Corporation, measures ths transit
time of acoustic or sonic energy propagating through air.
One model of such type of digitizer, called a "GRAPHBAR",
employs a pair of "point" microphones, having generally
circular receptivity patterns, mounted in spaced relation in
an elongated generally rectangular housing. The housing or
"bar" can be conveniently moved to a position adjacent an
area in which the position of a movable element, containing a
sound source, is to be digitized. The transit time of sound
traveling from the source to each microphone is used, in
conjunction with the speed of sound in air and known -
geometrical relationships, to compute the position of the
movable element.
In the described type of digitizer there is a region
WO91/10981 PCT/US91/004
~ 2 -~
adjacent the location of the transducers, sometimes referred
to as a "dead space", where it is difficult to determine the
position of the movable element with sufficient accuracy.
This is illustrated in conjunction with Fig. l which shows a
sonic digitizer that includes a bar 90 in which are mounted a
pair of spaced apart microphones 51 and 52. The microphones
are mounted near opposite ends of the bar and facing the area
lO to be digitized. [The size and shape of the area lO is
somewhat arbitrary and depends, inter alia, upon the
necessary accuracy of the digitizer readings.] The x and y
directions are as shown by the axes 59 in the diagram.
Consider the points l and 2, which are a distance d apart and
at respective distances L~ and L2 from microphone 51, and the
points 3 and 4 which are also a distance d apart and at
respective distances L3 and L4 from microphone 51.
Geometrical considerations dictate that the difference L4-L3
will be greater than the distance L2-L~. This makes it more
difficult to accurately determine the y coordinate location
of points near the bar 50. Therefore, a "dead space" ll
[whose specific size and shape are determined by desired
accuracy] is typically marked off and not used as part of the
area in which the position of the movable element is to be
accurately located. The "dead space" can be employed for a
function such as menu selection, if needed, which does not
require high accuracy in two dimensions.
In addition to the "dead space" being wasted in many
applications, the need to mark it off is a nuisance, and the
risk of inaccurate measurements in the "dead space", if the
movable element enters this area, is problematic.
It is among the objects of the present invention to
reduce the problems associated with digitizer "dead space"
and to generally improve the efficiency and compactness of
digitizer equipment. - - -
Graphical digitizer equipments, like most measuringequipments, are susceptible to errors caused by noise and
other factors. ~or example, through-the-air sonic digitizers
of the type described above are susceptible to extraneous
WO 91/10981 PCr/US91/00434
2 ~ 7
acoustic noise in the environment, and also to multipath
echoes of the sound energy employed by the digitizer
equipment itself. Electronic interference or other
intermittent pheno~ena can also lead to substantial
digitizing errors.
Since digitizers are typically utilized to measure and
store data points at a relatively hiqh acquisition rate, and
since the acquired data is often immediately used by a
companion system, the occurrence of occasional inaccurate
coordinate measure~ents, even grossly inaccurate ones, may
not be recognized at all, or until they cause a problem in
subsequent processing. The outputting of even occasional
incorrect coordinate data can be particularly undesirable for
certain applications. Further, when subsequent processing
involves averaging of acquired data points, a few grossly
inaccurate measurements can result in substantial errors in
averaged data that would otherwise be quite acceptable.
It is among the further objects of the present invention
to reduce or eliminate these type of problems in graphical
data digitizers and particularly, although not necessarily,
in sonic graphical digitizers.
The accurate determination of the transit time of the
acoustic energy between the transmitter and receiver
locations is critical to an accurate determination of the
position of the movable element. Typically, a timer is
provided for each receiver. All of the timers are started
when the acoustic energy is transmitted from the transmitter.
As the sound is received at each receiver, the timer
associated with that receiver is stopped. The transit times
to each receiver can then be computed from the time that
elapsed on each timer. Typically, each timer is a digital
counter which counts pulses from a digital clock generator,
and the arrival of acoustic wave energy at each microphone is
determined by continuously comparing the microphone output --
(e.g. an amplified and filtered version thereof) to a
predetermined threshold level. When the threshold level is
exceeded, the associated counter is turned off.
W091/10981 PCT/US91/~4~
- "I .
~ 4
In t ~ descri~ed type of system, a good source of
acoustic wave energy pulses is a spark gap which is energized
by triggering a circuit that delivers voltage pulses to a
pair of closely spaced electrodes which comprise the spark
gap. The trigger pulse for this circuitry is also
conventionally utilized to initiate the timer or timers that
are employed to measure the transit time of the acoustic wave
energy over an unknown distance to be determined. [As noted
above, the timers are subsequently terminated when the
acoustic wave energy is received at one or more respective
receivers. The measured elapsed time can be used for
determination of distance or, for pilot purposes, by
determination of the velocity of sound in air when the
transmitter to receiver distance is known.] The spark does
not occur immediately upon a~plication of the trigger signal
to the spark generation circuitry, so the timer(s) may be
initiated somewhat prematurely, resulting in an incorrect
elapsed time measurement. This would not necessarily be
problematic if one could determine the precise time
relationship between application of the trigger pulse and
occurrence of the spark, since suitable correction could then
be applied to the measured elapsed time. Applicant has
found, however, that such solution is generally not adequate,
since the time between the trigger and the actual spark can
vary considerably. There is a build-up time of the voltage
across the electrodes before a spark is produced (generally,
i at the output of the a transformer which is part of the spark
generation circuitry). The build-up may not be the same for
each spark to be generated and, also, the voltage at which a
spark is produced can vary over the life of the electrode
pair, and can also vary for different electrode pairs. This
means that the timing error will tend to vary and cannot be
readily accounted for by adding a predetermined timing
correction.
It is among the further objects of the present invention
to provide solution to the timing accuracy problem as set
forth.
~ .............................. . . .
WO 91/10981 PCrlUS91/00434
207~Dj'!7,
SUMMARY OF THE INVENTION
An aspect of the present inventlon is directed to an
apparatus for determining the position of a movable element.
In accordance with an embodiment of the invention, an
elongated housing is provided for positioning generally
adjacent an edge of an area in which the position of the
movable element is to be determined. The housing has a base
portion which contains a pair of spaced-apart transducers
that are mounted in the surface of the base portion and face
said area. An upper body portion of the housing is disposed
above the base portion and protrudes in cantilevered fashion
toward said area, so that the transducers are recessed from
said area beneath the protruding upper body portion of the
housing. Means are provided for determining the position of
the movable element from the respective transit times of
energy propagating in either direction between the movable
element and the pair of transducers.
In a form of the disclosed embodiment, the position-
determining means comprises electronic circuitry, and at
least a portion of the circuitry is contained within the
upper body portion. Also, in a form of the disclosed
embodiment, an additional transducer is mounted in the
recessed region beneath the protruding upper body portion,
for pilot purposes. Means are provided for determining the
transit time of energy propagating in either direction
between the additional transducer and at least one of said
pair of spaced-apart transducers. In this form of the
invention, the means for determining the position of the
movable element is responsive to both the respective transit
times of energy propagating in either direction between the
movable element and said pair of spaced-apart transducers and
the transit time of energy propagating in either direction
between the additional transducer and said at least one of
said pair of spaced-apart transducers.
The configuration set forth has the advantage of making
more efficient use of the "dead space" described above, and
WO93/10981 PCT/US91/~4
of pr~ ~d~ng a more compact di~itizer equipment. Also, the
configuration of the invention provides an advantageous
location for a pilot transducer.
An aspect of the present invention is directed to a
method and apparatus for more accurately determining the
transit time of acoustic energy travel between a transmitter
location and a receiver location. In a disclosed embodiment,
an electrode pair spark gap is provided at the transmitter
location, and an acoustic receiver is provided at the
receiver location. The spark-gap is energized to produce a
spark by coupling an electrical potential across the
electrode pair. Means are provided for sensing the
generation of a spark at the spark gap, and for generating an
initializing signal in response thereto. A timer is
initialized in response to the initializing signal. Means
are provided for detecting, at the receiver location, the
receipt of acoustic energy from the spark, and for generating
a terminating signal in response thereto. The timer is
terminated in response to the terminating signal, and the
time measured by the timer is indicative of the transit time
of acoustic energy travel between the transmitter and
receiver locations. In a preferred embodiment of this form of
the invention, the means for sensing the generation of a
spark at the spark-gap is operative to sense a current
coupled to the electrode pair. In this embodiment, the means
for energizing the spark gap includes: a transformer having
primary and secondary windings, the electrode pair being
coupled across the secondary winding; a capacitor coupled
cross the secondary winding; and means for applying a voltage
pulse to the primary winding. Also in this embodiment, the
means for sensing a current coupled to the electrode pair
comprises a transformer coupled to a conductor which couples
one of the :electrodes of the electrode pair to the secondary -
winding. This aspect of the present invention has application
to any technique or apparatus wherein it is desirable to
determine, accurately and consistently~ the transit time of
acoustic energy generated at a spark-gap; for example, two-
. ,. -:-~ z- . . -
W~91/10981 PCT/US91/~4~
7 ~ 7 7
dimensional acoustic digitizers, three-dimensional acoustic
digitizers, and one-dimensional acoustic distance or velocity
determination systems.
A further aspect of the present invention employs data
validation and screening based on the proximity of
sequentially measured data points. Another form of this
aspect of the invention detects noisy conditions manifested
in a pilot signal measurement, and discards data taken just
after (and/or, if desired, just before) detection of the
condition. In an illustrated embodiment of this aspect of the
invention, the position of a moveable element is determined
in two dimensions, but it will be understood that this aspect
of the invention has application to position determination in
one, two, or three dimensions, which utilizes any desired
num~er or configuration of transmitters and receivers.
Further features and advantages of the invention will
become more readily apparent from the following detailed
description when taken in conjunction with the accompanying
drawings.
.~ ,
BRIEF DESCRIPTION OF THE DRAWINGS
Fig. l is a diagram illustrating why the accuracy of
position determination is relatively low in a particular area
for a certain type of existinq digitizer equipment.
Fig. 2 is an elevational perspective view of an
apparatus in accordance with an embodiment of the invention.
Fig. 3 is schematic diagram, partially in block form, of
prior art circuitry which can be utilized in conjunction with
an embodiment of the invention.
Fig. 4 is an end perspective view of an apparatus in
accordance with an embodiment of the invention.
Fig. ~ is a side and bottom perspective view of an
apparatus in accordance with an embodiment of the invention.
Fig. 6 is an end view of another form of an apparatus in
accordance with an embodiment of the invention.
Fig. 7 is a schematic diagram of the spark generation and
wogl/l098l ~4~ i PCT/US91/004~
sensing circuitry in accordance with an embodiment of the
apparatus of the i~ention and which can be employed in
practicing an embodiment of the method of the invention.
Fig. 8, which includes Fig . s 8A and 8B placed one below
another, is a flow diagram of a routine for programming a
processor in accordance with practicing an embodiment of the
invention.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
~ eferring to Fig. 2, there is shown an apparatus in
accordance with an embodiment of a first aspect of the
invention for determining the position of an element movable
in a region located to one side of the apparatus 100 and
preferably, although not necessarily, within the dashed
region 10. The apparztus incudes an elongated housing 110
which is positioned generally adjacent an edge of the region
in which the position of a movable element 150 is to be
determined. The housing 110 has a base portion 111 which
contains a pair of spaced apart transducers 20 and 30 that
are mounted in a surface 118 of the base that faces the
region 10. The housing 110 has an upper body portion 112
which is disposed above the base portion and which protrudes,
in cantilevered fashion, toward the region 10, so that the
transducers 20, 30 are recessed, with respect to the region
10, beneath the cantilevered upper body portion 112 of the
housing 110. In the present embodiment, the front surface
113 of the upper body portion 112 i5 approximately above the
edge of the region 10. In this manner, if desired, the upper
body portion can serve to prevent a suitably configured
movable element 150, such as a stylus, pen, finger, cursor,
or the like, from enterïng the area ll beneath the protruding
upper body portion 112. As described in the background
portion hereof, the area directly adjacent the transducers in
an existing bar-type digitizer is generally not utilized
since the determination of position in this area may lack
sufficient accuracy. The present invention has the ~dvantage
WO 91/10981 PCl'tUS9 1 /0043'S
9 2i~4~77
of eliminating the need for markiny off a "dead space"
adjacent the digitizer apparatus, as well as preventing
unintended positioning of the movable element in the "dead
space". Further toward this end, and as shown in Fig. 4, a
suitable sound-transmitting screen 140 can be provided to
enclose part or all of the region beneath the upper body
portion 112. The screen may be, for example, of plastic
mesh, and will preferably permit free circulation of air.
A further ad~antage of the present invention is that the
upper body portion can, if desired, be utilized to contain a
portion of the electronics used in the position determination
function. In this manner, the so-called dead space is not
wasted, and the housing 110 can have a compact configuration.
In one form of the embodiment of Fig. 2, the transducers
20 and 30 are acoustic receivers, such as point microphones,
and the movable e!ement 150 is a stylus (or cursor puck, or
other suitable device), which contains a transducer for
producing acoustic wave energy. Techniques for determining
the position of a movable element sound emitter with respect
to a pair of receivers, such as point microphones, are well
known in the art. Briefly, however, and as illustrated in
Fig. 3, the travel time duration is determined by circuitry
40, shown for convenience in dashed line to the rear of bar
90, which comprises a left counter 42, associated with the
left microphone 20, a right counter 43 associated with the
right microphone 30, a clock 44, and a spark generator
circuit 41. Coincident with generation of the spark at
movabl~ element 150, the counters 42 and 43 are enabled by a
gating signal from the spark generator circuit to begin
counting pulses from clock 44. Upon initial reception of the
sound wavefront, the microphones 20 and 30, which generally
receive the wavefront at different times, produce output
voltages which are coupled to high gain band pass amplifiers
21 and 31, respectively. The spark shock wave produces a
fast rise time electrical impulse upon impinging on the
microphone surface, and the band pass amplifiers allow only
the fast rise time portion of the electrical pulse to pass
WO91/1098~ J~I PCT/US91/004
while blocking out noise signals outside the band. To insure
rapid operation, the amplifiers include threshold
discriminators which provide an output pulse with a steep
leading edge in response to the input thereto exceeding a
predetermined level. The amplifier outputs are op~rative to
disable the counters 42 and 43 and also to read out the
respective counts which are indicative of the travel times
between the sound source on the movable element and the
microphones. The respective distances can then be computed,
in known manner, by multiplying the travel times by the
velocity of sound in air. This can be implemented, for
example, by computing module or processor 200, or by any
suitable dedicated or general purpose processor.
Fig. 5 illustrates an embodiment of a further feature of
the first aspect of the invention which utilizes a third fixed
transducer 123 mounted on the housing 110 for the purpose of
obtaining a velocity-representative signal that is used in
deriving more accurate digitizer position determinations. As
is well known in the art, the speed of sound through air
varies substantially with the temperature of the air, and
acoustic digitizers can utilize a measurement between fixed
distances, sometimes called a pilot measurement, to obtain
temperature compensated digitizer outputs. The fixed distance
can be obtained, for example, by placing the movable element
at a known position before taking a pilot measurement. This
has the disadvantage of requiring a time-consuming manual
operation. Also, since subsequen.t measurements are taken
after significant time has passed, changes in conditions can
occur, which would reduce the effectiveness of the pilot
measurement. Another known technique is to utilize a second
sound source (or receiver, if the fixed transducers are
sources) which is at a fixed position with respect to the
receivers. However, this gives rise to the problem of where -
to place the further fixed transducer so that it will not be
obtrusive, and so that it will not interfere in any way with
operation of the movable element of digitizer. In the Fig. 5
embodiment, a further fixed transducer 123 is positioned on
.~ - .
- - . - ~
WO 91/10981 PCr/US91/00434
1 1 2 ~ 7 ~ f~ ~ 7
the bottom wall of the protruding upper body portion 112, and
this solves the aforementioned problems while, again, making
use of the so-called "dead space" and not interfering with
normal digitizer operation. In operation (see also Fig. 3), a
spark gap can be provided as transducer 123. A spark
generator (such as 41) can energize the spark gap 123 and
clock pulses (such as from clock 44) are counted by a counter
(e.g. 43 coupled to microphone 30) until the counter is
disabled by arrival of the sound wavefront at microphone 30.
The count represents the transit time of the sound wavefront.
The speed of sound in the present air environment can then be
computed by dividing the known distance (between fixed source
123 and microphone 30) by the obtained transit time. This
speed of sound can then be utilized in the above-referenced
distance computations for the movable element. The pilot
measurements can also be used for determining the validity of
position measurement data, as will be described hereinbelow.
It will be understood that the pilot measurements can be made
as frequently as desired.
The embodiment of Fig. 2 was illustrated in terms of a
position determining apparatus in which the movable element
includes a sound source, and the transducers 20 and 30 are
sound~receivers. It will be understood, however, that, if
desired, either or both of the transducers 20 and/or 30 can
be utilized to transmit acoustic energy. In such case, the
movable element can be utilized as a receiver, thereby
reversing the mode of operation which was first described.
As is known in the art, the transmitters can be sequentially
energized, and the distance be~ween each transmitter and the
receiver in the movable element can be computed in the manner
previously described. From this information, and known
trigonometric relationships, the position of the movable
element can be determined. In a still further variation, the
movable element can be a passive reflector'of acoustic
energy.- In this regard, see, for example, U.S. Patent No.s
4,012,588 or 4,124,838, assigned to the same assignee as the
present application. In such case, one or both of the
WO91/10981 PCT/US91/~4
~ 12
transducers 30, 40 could be used as a transmitter as well as
a receiver. If desired, a separate transmitter can also be
employed. It will also be understood that the pilot
transducer 123 can be a receiver whén the transducers 20
and/or 30 are transmitters. Further, the shape contours of
the housing and the protruding upper body portion can be
varied to some degree while retaining the indicated
advantages of the invention. Also, additional structure can
be provided for support, balance, or other purposes. For
example, in Fig. 6 a base panel 105 is provided and, if
desired, the front thereof can be used for a menu selection
function.
Fig. 7 illustrates an embodiment of the spark generator
circuit 4l as improved in accordance with an aspect of the
present invention, and which can be used in practicing an
embodiment of a method in accordance with the invention. In
the embodiment of Fig. 7, a supply voltage, V~, is utilized to
charge a capacitor Cl via a resistor R1. The capacitor has a
discharge path through the primary winding of a transformer T
and a silicon controlled rectifier (labeled SCR), when the
SCR is conductive. As in known in the art, trigger pulses
are applied, at appropriate times, to the trigger the gate
electrode g of the SCR to render the SCR conductive and cause
a pulse of relatively high voltage across the transformer
secondary winding. When the SCR turns off, the capacitor can
again be charged and awaits the next trigger pulse. The
circuit, as just described, is known in the art, and it can
be noted that prior art systems typically also utilize the
trigger signal, or a signal derived therefrom, to initialize
the counters 42 and 43, as first described above.
In the present embodiment, the secondary winding of the
transformer Tl is coupled, via a filter 210 and cable 220,. to
a spark gap electrode pair 225 which is illustrated as being
at the tip of a stylus 150 (as in Fig.s 2 and 3).- The filter~.
210 comprises series resistors R2 and R3, and a capacitor C2
in parallel with the spark gap. In this embodiment, the
current to the spark gap is sensed, without conductive
WO91/10981 PCT/~S91/004~
13 2137~G~77
coupling, by utilizing a transformer 250. In an operating
embodiment hereof, a twin-hole balun core was employed for
this purpose. One of the conductors that is coupled to cable
220 is passed through a hole of the balun core 250. A
further conductor 260 is passed through the other hole of the
balun core. One end of conductor 260 is coupled to ground
reference potential, and the other end is coupled, via a
diode D1 and a resistor R~, to the gate electrode of a
field-effect transistor Q1- The gate electrode of Q1 is alco
coupled, via resistor R5, to ground reference potential. In
the present embodiment, the drain electrode of Q1 is coupled
to a positive bias voltage V via a resistor R6, and the
source electrode of Q1 is coupled to ground reference
potential. An output 270, which is taken at the drain
electrode of transistor Q1, is coupled to the enable inputs of
counters 42 and 43 (as in Fig. 3).
In operation, the network comprised of R2, R3 and C2
forms low pass filter 210, which limits the transformer
secondary current at breakdown [i.e., when there is arcing
across the spark gap electrode pair]. C2 discharges very
guickly when the arc is initiated, and a very short steep
current pulse flows from C2 into the cable 220 at the onset of
the arc. The magnitude of the pulse depends on the value of -~
C2, the breakdown voltage, and the speed of breakdown. The
occurrence of this current pulse indicates, with good
precision, the time at which the arc occurs and it is sensed,
in the present embodiment, to develop a signal that is
consistently related to the time of onset of the acoustic
wave energy caused by the spark. The current pulse in the
conductor passing through transformer 250 induces a
corresponding pulse in conductor 260. This signal, applied
to the gate electrode of Q1, turns Ql on and causes the output
voltage at 270 to go from V to substantially ground reference
potential for as long as Q1 is on. [Of course, if a
positive-going signal rather than a negative-going signal is
desired for enabling the clocks 42, 43, the output on 270 can
be suitably converted, or a suitable circuit which directly
W091/10981 ~ PCT/US91/0
~ ~ 14
generates a positive-going signal can be employed.]
The very short pulse from the secondary of transformer
250 charges the electrode capacitances of the field-effect
transistor Qi which "stretch" the output while discharging
through Rs~ [If these capacitances are not adequate to
sufficiently "stretch" the pulse, a small capacitance can be
added between ground reference and the junction of D, and R4.
This will, however, subject the diode to a higher reverse
voltage at the end of the pulse.] The series resistor, R4,
limits the peak charging current and prevents the high peak
voltage from appearing at the gate electrode. The diode D
may be, for example, a Schottky-barrier qiode with a fast
reverse recovery time. It will be understood that other
suitable circuits could be used for detecting the spark
onset.
In an aspect of the invention to be described next, two
types of digitizer data validity determinations are used
together, although it will be understood that either form can
be used to advantage without the other.
Referring to Fig. 8, there is shown a flow diagram of a
routine for control of the processor 200 in accordance with
an embodiment of the invention. The processor may comprise,
for example, an IBM-PS2, together with conventional associated
memory, timing, input/output and display functions (not
shown). The diamond 802 represents inquiry as to whether the
data validity mode is active. If not, the routine is exited.
The routine can then be re-entered, such as after a suitable
interrupt resulting from an operator selection or periodic
inquiry from another routine. It is assumed that the current
routine will generally be implemented when a continuous data
type of mode is being used at the digitizer, although it will
be understood that the routine can be employed in conjunction--
with any data mode. The routine is illustrated as being
entered via a suitable interrupt (e.g., when a continuous run
mode is entered), or via operator control. If the data valid 1 -
test mode is operative, inquiry is made (diamond 805) as to
whether the mode was just rendered operative. If so, the
WO9~/10981 PCTIUS91/004~
2 0 7~ 0 i 7
block 808 is entered, this block representing the clearing of
registers used for temporary storage of data points in
conjunction with the data validity de~erminations and the
clearing of any previously set flags. An initializing status
flag is then set, as represented by the block 810. The
decision diamond 813 is then entered (and is also entered
from the "no" output branch of diamond 805), and inquiry is
made as to whether the data currently being sought is pilot
data or movable element data (i.e., stylus, cursor, or any
movable element data). [For ease of explanation, movable
element data will be referred to as stylus data for this part
of the description.] In the present illustrated embodiment,
it is assumed that pilot and stylus data are alternately
obtained, under control of processor 200, such as by
alternately energizing the stylus 150 and the pilot
transducer 123 of Fig. 5. Also, in the illustrated
embodiment it is assumed that no stylus data point will be
deemed valid until both types of validity checks described
herein, viz. validity based on the pilot data and validity
based on the relative positions of adjacent stylus data, are
satisfied. If pilot data is currently being sought, the data
is awaited (diamond 815, loop 816, or arrows 817 if
interrupted and returned during the wait) and, when the
measurament data has been received the counter time of the
counter used for the pilot (43 in Fig. 2) is tested (diamond
820) to determine whether it is in a predetermined acceptable
range. The range of travel times may be determined, for
example, from the travel time of sound over the fixed pilot
distance at the lowest and highest expected operating air
temperatures, for a through-the-air digitizer. If the pilot
measurement is within acceptable range, a pilot acceptability
indicator bit is set high (block 825), whereas if the
measurement is outside the range, the pilot acceptability bit
is set low (block 827). The diamond 813 is then re-entered.
If the inquiry of diamond 813 indicates that stylus
measurement data is currently being sought, the data is
awaited, as represented by diamond 830, loop 831, and arrows
WO91/10981 PCT/US91/~4
~ 16
832 which, again, indicate that interrupts can be used for
performance of other functions during the wait. When the
stylus measurement data arrives, the counts of counters 42
and 43 are used, in known fashion, to compute the stylus
position, as represented by the block 835. Inquiry is then
made (diamond 837) as to whether the initial status flag is
set. If so, inquiry is made (diamond 840) as to whether the
pilot acceptability bit is high. If so, the data coordinate
values are stored in a previous point register (block 842),
the initial status flag is turned off (block 845), and the
diamond 813 is re-entered. If, however, the pilot
acceptability bit is low, the computed stylus position is not
stored, and the diamond 813 is re-entered directly. [Thus,
in the absence of an acceptable pilot,the initial status is
maintained, and no data will be read-out or stored for
comparison.]
In the situation where the initial status flag is not
set (the "no" branch of diamond 837), inquiry is again made
(diamond 850) as to the pilot acceptability bit status. If
the pilot acceptability bit is low, the previous point
register is cleared (block 852), the initial status flag is
set (block 853), and the diamond 813 is re-entered. [Thus,
in the absence of an acceptable pilot, the initial status is
re-established and no data will be read-out or stored for
comparison.] If the pilot acceptability bit is high, the
block 860 is entered, this block representing the computation
of the distance between the current stylus data point and the
previous data point stored in the previous point register.
If the current point coordinates are (xc, Yc) and the previous
point coordinates are (xp, yp), then the computed distance
will be
D = ~ (X - X ~2 +(y _ y 2~ ~l/2
[For a three dimensional system, there will be a similar z
term.] Inquiry is then made (diamond 862) as to whether the
distance D is less than or equal to the maximum acceptable
WO91/10981 P~T/US91/004~
17 2~7~77
movement distance, desiynated Do. The acceptable region
around the prior point (i.e., a circle of radius Do for this
embodiment) can be 2reselected and/or made operator
adjustable, and depends on the maY~imum expected stylus
movement in the time interval between sequentially adjacent
stylus data points. If D is within acceptable range, the
current point is read-out (and/or placed on a valid data
list), as represented by the block 865. Also, the current
data point replaces the one stored in the previous point
register (block 867). If D is not within the acceptable
range, the previous point register is cleared (block 870),
the initial status flag is set (block 873), and diamond 813 is
re-entered. ~hus, after an invalid data point, the
validation process starts from scratch in the present
embodiment.
The structures of Fig.s 1-6 illustrated in terms of a
position determining apparatus in which the movable element
includes a sound source, and the transducers 20 and 30 are
sound receivers. It will be understood, however, that, if
desired, either or both of the transducers 20 and/or 30 can
be utilized to transmit acoustic energy. In such case, the
movable element can be utilized as a receiver, thereby
reversing the mode of operation which was first described.
As is known in the art, the transmitters can be sequentially
energized, and the distance between each transmitter and the
receiver in the movable element can be computed in the manner
previously described. From this information, and known
trigonometric relationships, the position of the movable
element can be determined. In a still further variation, the
movable element can be a passive reflector of acoustic
energy. In such case, one or both of the transducers 30, 40
could be used as a transmitter as well as a receiver. If
desired, a separate transmitter can also be employed. It
will also be understood that the pilot transducer 123 can be
a receiver when the transducers 20 and/or 30 are
transmitters. The invention is applicable to all of these
situations, as well as to other types of digitizers.
WO91/10981 PCT/US91/~
4Q~ 8
In the just illustrated embodiment, current points are
compared with previously acquired points, and either outputted
or not. However, if desired, after comparison against a
current point, a previously stored point can be read out or
outputted, or a combina~ion of both techniques could be
utilized. Also, it will understood that comparison of three
or more points could also be utilized to localize the
moveable element and determine the region of acceptability.
If desired, the allowable region in which adjacent points
should be located can be non-symmetrical. For example, if
the direction of motion is computed from previous points,
more leeway could be permîtted for the next point in the
general direction of motion, consistent with the physics of
hand movement. Further along these lines, the velocity (rate
of change of position) of motion could also be used in
determining the acceptable region for an adjacent point.
Also, it will be understood that while the disclosed
embodiment uses a computation of distance moved based on
computed coordinates, the sequentially obtained slant ranges
could be directly compared, if desired. Further, it will be
understood that suitable indices can be associated with
received points to keep track of points which are read out
and/or those which are screened out. Visual and/or audio
indicators can also be generated whenever invalid data is
detected.
.
