| United States Patent Application |
20130002600
|
| Kind Code
|
A1
|
|
McCracken; David Harold
|
January 3, 2013
|
TOUCH SENSITIVE DEVICE ADAPTIVE SCALING
Abstract
A method for performing adaptive scaling in a touch sensitive device
including a touch pad is provided. The method includes obtaining a
trajectory of touch positions from the touch pad; setting a first scaling
factor; comparing an acceleration factor to a deceleration factor, and:
setting a second scaling factor to the acceleration factor if the first
scaling factor is lower than the acceleration factor when the
acceleration factor is greater than the deceleration factor; setting the
second scaling factor to the deceleration factor if the first scaling
factor is greater than the deceleration factor when the acceleration
factor is lower than or equal to the deceleration factor; and updating
the trajectory with a new touch position provided by the touch pad and a
scaling factor set to the second scaling factor. A touch sensitive device
coupled to a display for use with the above method is also provided.
| Inventors: |
McCracken; David Harold; (Aptos, CA)
|
| Serial No.:
|
198526 |
| Series Code:
|
13
|
| Filed:
|
August 4, 2011 |
| Current U.S. Class: |
345/174 ; 345/173 |
| Class at Publication: |
345/174 ; 345/173 |
| International Class: |
G06F 3/044 20060101 G06F003/044; G06F 3/041 20060101 G06F003/041 |
Claims
1. A method for performing adaptive scaling in a touch sensitive device
comprising a touch pad having a sensing range and a display having a
display range comprising: obtaining a trajectory of touch positions from
the touch pad; setting a first scaling factor; comparing an acceleration
factor to a deceleration factor, the acceleration and deceleration
factors related to an acceleration and deceleration of the touch
positions along the obtained trajectory and: setting a second scaling
factor to the acceleration factor if the first scaling factor is lower
than the acceleration factor when the acceleration factor is greater than
the deceleration factor; setting the second scaling factor to the
deceleration factor if the first scaling factor is greater than the
deceleration factor and the acceleration factor is lower than or equal to
the deceleration factor; and updating a trajectory on the display using a
new touch position provided by the touch pad and the second scaling
factor.
2. The method of claim 1 further comprising: obtaining at least an
inflection point in the trajectory of touch positions; obtaining a speed
of motion of a touch at each point relative to one of a first trajectory
point or a previous inflection point; and setting the acceleration factor
proportional to the speed of motion.
3. The method of claim 2 wherein obtaining at least an inflection point
comprises: selecting a starting point from the trajectory; selecting a
subsequent point in the trajectory separated from the starting point by a
distance greater than a pre-selected distance; obtaining a direction
between the starting point and the selected point; selecting the
inflection point when the obtained direction is different from a previous
direction by an amount greater than a pre-selected direction change.
4. The method of claim 3 wherein selecting the inflection point comprises
comparing the direction between the starting point and the selected point
using a pre-selected coarse direction.
5. The method of claim 1 wherein comparing an acceleration factor to a
deceleration factor further comprises the steps of: obtaining a speed of
motion of a touch on the touch sensitive device; computing the
acceleration factor proportional to the speed of motion; obtaining an
envelope of the trajectory; obtaining a measure for the envelope of the
trajectory; and computing the deceleration factor proportional to the
measure for the envelope.
6. The method of claim 1 wherein setting a first scaling factor comprises
obtaining the first scaling factor from a sensing range to cover about
50% of a display range.
7. The method of claim 1 wherein the obtaining a trajectory of touch
positions further comprises filtering a jitter motion of a touch on the
touch sensitive device to provide a jitter free trajectory.
8. The method of claim 7 wherein filtering a jitter motion comprises:
selecting a first point from the trajectory; selecting a number of points
from the trajectory less than a pre-selected maximum count; obtaining a
filtered position from the selected points; wherein the selected points
are closer to the first point than a pre-selected range.
9. The method of claim 8 wherein obtaining a filtered position comprises
obtaining an average of the positions of the selected points.
10. The method of claim 5 wherein obtaining the envelope of the
trajectory comprises: selecting a plurality of trajectory points in a
buffer; finding a circumscribing polygon for the plurality of points;
updating the buffer with a new point in the trajectory.
11. The method of claim 10 wherein the buffer is a circular buffer.
12. The method of claim 10 wherein finding a circumscribing polygon
comprises: selecting a first maximum coordinate in a first direction and
a second maximum coordinate in a second direction for the points in the
buffer; selecting a first minimum coordinate in the first direction and a
second minimum coordinate in the second direction for the points in the
buffer; finding four edges of the circumscribing polygon using the first
maximum coordinate, the second maximum coordinate, the first minimum
coordinate, and the second minimum coordinate.
13. The method of claim 10 wherein the measure for the envelope comprises
the perimeter of the circumscribing polygon.
14. The method of claim 12 wherein the measure for the envelope
comprises: a first term subtracting the first maximum and the first
minimum; and a second term subtracting the second maximum and the second
minimum.
15. The method of claim 1 further comprising: detecting a distance
between a touch and an edge boundary of the touch pad; obtaining a
scaling adder; obtaining a third scaling factor by adding the scaling
adder to the second scaling factor; updating the trajectory on the
display using a new touch position provided by the touch pad and the
third scaling factor.
16. The method of claim 15 wherein obtaining the scaling adder comprises:
increasing the scaling adder when the distance between the touch and the
edge boundary is less than a pre-selected distance.
17. The method of claim 15 wherein obtaining the scaling adder comprises:
increasing the scaling adder when a user intent is to continue a motion
towards the edge of the touch pad.
18. The method of claim 17 wherein the user intent is determined by:
measuring a touch speed; measuring a change of direction in the touch
speed; and measuring a touch strength on the touch pad.
19. The method of claim 1 further comprising: detecting a distance
between a touch and an edge boundary of the touch pad; and determining
user intent to avoid edge rollback.
20. The method of claim 19 wherein determining user intent comprises
measuring a jitter in the touch motion; further wherein the user intent
is determined to be `continued motion` when the measured jitter is
smaller than a pre-selected value.
21. The method of claim 19 wherein determining the user intent comprises
measuring a strength of a touch signal in the touch pad; and the user
intent is determined to be `continued motion` when the measured strength
decreases monotonically as the touch approaches the edge boundary.
22. The method of claim 21 wherein measuring the strength of a touch
signal comprises a capacitance measurement.
23. The method of claim 1 further comprising: measuring a strength of a
touch signal in the touch pad; and determining user intent to avoid an
untouch jump using the measured strength.
24. The method of claim 23 wherein determining user intent comprises:
obtaining a change of strength of a touch signal; and determining user
intent to continue a touch when the change of strength of the touch
signal is smaller than a pre-selected strength change value.
25. The method of claim 23 wherein determining user intent comprises:
obtaining a change of speed of motion of a touch on the touch sensitive
device; and determining user intent to continue a motion when the change
is smaller than a pre-selected speed change value.
26. The method of claim 23 wherein determining user intent comprises:
determining user intent to untouch when a condition occurs, the condition
selected from the group consisting of a change of strength of a touch
being larger than or equal to a pre-selected strength change value, and a
change of speed of motion being larger than or equal to a pre-selected
speed change value.
27. The method of claim 23 wherein measuring the strength of a touch
signal comprises a capacitance measurement.
28. A method for scaling a movement on a sensitive pad to a movement on a
display comprising: obtaining a trajectory from the sensitive pad;
setting a first scaling factor; obtaining a speed of motion from the
trajectory; obtaining a measure for a short-range movement on the
sensitive pad; computing an acceleration factor proportional to the speed
of motion; computing a deceleration factor proportional to the measure
for a short-range movement; comparing the deceleration factor to the
first scaling factor, and: setting a second scaling factor to the
acceleration factor when the first scaling factor is lower than the
acceleration factor when the deceleration factor is greater than or equal
to the first scaling factor; setting the second scaling factor to the
deceleration factor when the first scaling factor is greater than or
equal to the acceleration factor when the deceleration factor is lower
than the first scaling factor; setting the second scaling factor to a
weighted average of the acceleration factor and the deceleration factor
when the first scaling factor is greater than the deceleration factor and
the acceleration factor is greater than the first scaling factor; and
updating a trajectory on the display with the second scaling factor.
29. The method of claim 28 wherein obtaining a measure for the
short-range movement comprises: obtaining an envelope of the trajectory;
obtaining a measure for the envelope of the trajectory; and computing the
measure for the short-range movement proportional to the measure for the
envelope.
30. The method of claim 28 wherein obtaining a trajectory from the
sensitive pad further comprises filtering a jitter motion of a touch on
the sensitive pad to provide a jitter free trajectory.
31. The method of claim 28 wherein setting a first scaling factor
comprises obtaining the first scaling factor from a sensing range to
cover about 50% of a display range.
32. The method of claim 28 further comprising obtaining at least an
inflection point in the trajectory from the sensitive pad and obtaining
the speed of motion at each point relative to one of a first trajectory
point or a previous inflection point.
33. A method for performing adaptive scaling in a touch sensitive device
comprising a touch pad having a sensing range and a display having a
display range comprising: obtaining a trajectory of touch positions from
the touch pad; setting a direction factor; adjusting a first scaling
factor in a first direction and a second scaling factor in a second
direction using the direction factor and a coarse direction of the
trajectory; and updating the trajectory with a new touch position in the
first direction using the first scaling factor and in the second
direction using the second scaling factor.
34. The method of claim 33 wherein adjusting the first scaling factor and
the second scaling factor comprises: obtaining an envelope of the
trajectory; and obtaining a first measure in the first direction of the
envelope and a second measure in the second direction of the envelope.
35. A method for performing adaptive scaling in a touch sensitive device
comprising a touch pad having a sensing range and a display having a
display range comprising: obtaining a trajectory of touch positions from
the touch pad; obtaining a first value proportional to a long range
performance; obtaining a second value proportional to a short range
performance; adjusting a scaling factor using a difference between the
first value and the second value; and updating a trajectory on the
display with a new touch position using the scaling factor.
36. The method of claim 35 wherein the long range performance is obtained
by using at least an inflection point in the trajectory of touch
positions and obtaining a speed of motion of a touch at each point in the
trajectory.
37. The method of claim 35 wherein the short range performance is
obtained by: obtaining an envelope of the trajectory; and obtaining a
measure for the envelope of the trajectory.
38. A method for performing adaptive scaling in a touch sensitive device
comprising a touch pad having a sensing range and a display having a
display range comprising the steps of: obtaining a trajectory of touch
positions from the touch pad; setting a first scaling factor; obtaining
an acceleration factor proportional to a speed of motion of a touch;
obtaining a deceleration factor proportional to a measure of an envelope;
identifying the location of a target object in the display; increasing
the acceleration factor when the trajectory overlaps the target object;
decreasing the acceleration factor when the trajectory ceases to overlap
the target object; setting a second scaling factor to the acceleration
factor when the acceleration factor is greater than the deceleration
factor; setting the second scaling factor to the deceleration factor if
the first scaling factor is greater than the deceleration factor when the
acceleration factor is lower than or equal to the deceleration factor;
and updating the trajectory with a new touch position provided by the
touch pad and the second scaling factor.
39. A touch sensitive device coupled to a display, the touch sensitive
device having a touch pad and a controller comprising: a processor
circuit coupled to receive data from the touch pad, wherein the processor
circuit obtains a touch location from data provided by the touch pad; a
memory circuit coupled to receive and store the touch location from the
processor circuit and form a trajectory from a plurality of touch
locations, wherein: the processor circuit obtains an instantaneous speed
and an envelope having a measure from the trajectory stored in the memory
circuit; and the controller provides a signal to the display to move an
indicator to a position on the display; and the position on the display
is obtained by the processor circuit using the touch location and a
scaling factor computed using the instantaneous speed and the envelope
measure.
40. The touch sensitive device of claim 39 wherein the processor circuit
executes instructions stored in the memory circuit for performing a
method comprising: obtaining a trajectory of touch positions from the
touch pad; setting a first scaling factor; comparing an acceleration
factor to a deceleration factor; setting a second scaling factor to the
acceleration factor if the first scaling factor is lower than the
acceleration factor when the acceleration factor is greater than the
deceleration factor; setting the second scaling factor to the
deceleration factor if the first scaling factor is greater than the
deceleration factor when the acceleration factor is lower than or equal
to the deceleration factor; and updating a trajectory on the display
using a new touch position provided by the touch pad and the second
scaling factor.
41. The touch sensitive device of claim 40 wherein comparing an
acceleration factor to a deceleration factor comprises: obtaining a speed
of motion of a touch on the touch sensitive device; obtaining an envelope
of the trajectory; obtaining a measure for the envelope of the
trajectory; computing the acceleration factor proportional to the speed
of motion; and computing the deceleration factor proportional to the
measure for the envelope.
42. The touch device of claim 40 wherein the obtaining a trajectory of
touch positions further comprises filtering a jitter motion of a touch on
the touch sensitive device to provide a jitter free trajectory.
43. The touch device of claim 40 wherein the setting a first scaling
factor comprises obtaining the first scaling factor from a sensing range
to cover about 50% of a display range.
44. The touch device of claim 40 further comprising obtaining at least an
inflection point in the trajectory of touch positions and obtaining the
speed of motion at each point relative to one of a first trajectory point
or a previous inflection point.
45. The touch device of claim 39 wherein the touch pad comprises a
capacitively coupled touch sensor.
46. The touch device of claim 39 wherein the touch pad comprises an
optically coupled touch sensor.
47. The touch device of claim 39 wherein the touch pad provides data to
the processor circuit from a touch event produced by a finger.
48. The touch device of claim 39 wherein the touch pad provides data to
the processor circuit from a touch event produced by a touch device
selected from the group consisting of a finger, a stylus, and a pen
device.
Description
CROSS-RELATED APPLICATIONS
[0001] This application claims priority to U.S. Provisional Patent
Application No. 61/504,002 filed Jul. 1, 2011, entitled "Touch Sensitive
Device Adaptive Scaling" by David Harold McCracken, the disclosure of
which is incorporated by reference in its entirety here for all purposes.
[0002] This application is related to U.S. patent application entitled
"Touch Device Gesture Recognition" (Attorney Docket No. 70107.327) by
David Harold McCracken, assigned to Integrated Device Technology, Inc.
filed concurrently with the present disclosure on Aug. 4, 2011, and which
is incorporated herein by reference in its entirety for all purposes.
This application is also related to U.S. patent application Ser. No.
13/154,227, filed on Jun. 6, 2011, entitled "Differential Capacitance
Touch Sensor" by David Harold McCracken, assigned to Integrated Device
Technology, Inc. incorporated herein by reference in its entirety for all
purposes.
BACKGROUND
[0003] 1. Technical Field
[0004] Embodiments described herein generally relate to the field of touch
sensitive devices that transfer position data to a display device. More
particularly, embodiments disclosed herein relate to methods to adjust
scaling in touch sensitive devices according to user intent.
[0005] 2. Description of Related Art
[0006] In the field of touch sensitive devices the display typically has a
surface area that may be much larger than the sensitive area in the touch
sensitive pad. To adjust for the different sizes between the sensitive
pad and the display a scaling factor is used to transform a position on
the pad to a position on the display. As the user moves a cursor over the
display, it may need precise positioning to reach a specific target on
the display. In other occasions, the user may need to translate a cursor
or a pointer across a large portion of the display, in a long, fast slide
motion. While precise positioning may use a small scaling factor for the
translation from the touch sensitive pad to the display, a long slide
motion may use a larger scaling factor. This presents the problem of
adjusting the scale of the motion translation according to the user
needs. Furthermore, the adjustment is preferably performed in a timely
manner, to avoid sluggishness in the response, also allowing the user
some range for a jittery, indecisive motion.
[0007] Current state-of-the-art solutions to the problem of scaling in
touch sensitive devices implement a speed-based correction factor.
According to these solutions, the scaling factor is adjusted by measuring
the acceleration of a finger motion in the touch pad. Thus, for example,
a larger scaling factor resulting in larger display movements is
implemented for faster movement in the touch pad. Likewise, a smaller
scaling factor resulting in more precise display movements is implemented
for a decelerating movement in the touch pad. However, a speed-based
approach fails to accurately follow a user's intent in many
circumstances. For example, a fast moving touch may incorrectly suggest
that the user intends to move a cursor on the display further away.
Another problem with a speed-based approach is that the instantaneous
speed of a single stroke or slide may vary substantially from the start
point to the end point. Unless there is a mechanism to interpret sudden
changes in speed, the device may become sluggish if simple averaging of
the speed is used. These problems are exacerbated for systems using small
input touch devices, since accelerated movements are more frequent in
these systems.
[0008] What is needed is a touch sensitive device and a method for using a
touch sensitive device that accurately and rapidly adjusts the
translation scale according to user intent.
SUMMARY
[0009] According to embodiments disclosed herein, a method for performing
adaptive scaling in a touch sensitive device including a touch pad having
a sensing range and a display having a display range may include:
obtaining a trajectory of touch positions from the touch pad; setting a
first scaling factor; comparing an acceleration factor to a deceleration
factor, and: setting a second scaling factor to the acceleration factor
if the first scaling factor is lower than the acceleration factor when
the acceleration factor is greater than the deceleration factor; setting
the second scaling factor to the deceleration factor if the first scaling
factor is greater than the deceleration factor when the acceleration
factor is lower than or equal to the deceleration factor; and updating
the trajectory with a new touch position provided by the touch pad and
the second scaling factor.
[0010] According to embodiments disclosed herein, a method for scaling a
movement on a sensitive pad to a movement on a display may include
obtaining a trajectory from the sensitive pad; setting a first scaling
factor; obtaining a speed of motion from the trajectory; obtaining a
measure for a short-range movement on the sensitive pad; computing an
acceleration factor proportional to the speed of motion; computing a
deceleration factor proportional to the measure for a short-range
movement; comparing the deceleration factor to the first scaling factor,
and: setting a second scaling factor to the acceleration factor if the
first scaling factor is lower than the acceleration factor when the
deceleration factor is greater than or equal to the first scaling factor;
setting the second scaling factor to the deceleration factor if the first
scaling factor is greater than or equal to the acceleration factor when
the deceleration factor is lower than the first scaling factor; setting
the second scaling factor to a weighted average of the acceleration
factor and the deceleration factor if the first scaling factor is greater
than the deceleration factor and the acceleration factor is greater than
the first scaling factor; and updating the trajectory with the second
scaling factor.
[0011] Further according to embodiments disclosed herein a method for
performing adaptive scaling in a touch sensitive device including a touch
pad having a sensing range and a display having a display range may
include: obtaining a trajectory of touch positions from the touch pad;
setting a direction factor; and adjusting a first scaling factor in a
first direction and a second scaling factor in a second direction using
the direction factor and a coarse direction of the trajectory. The method
may further include updating the trajectory with a new touch position in
the first direction using the first scaling factor and in the second
direction using the second scaling factor.
[0012] A method for performing adaptive scaling in a touch sensitive
device including a touch pad having a sensing range and a display having
a display range according to embodiments disclosed herein may include:
obtaining a trajectory of touch positions from the touch pad; obtaining a
first value proportional to a long range performance; obtaining a second
value proportional to a short range performance; adjusting a scaling
factor using a difference between the first value and the second value;
and updating a trajectory on the display with a new touch position using
the scaling factor.
[0013] A method for performing adaptive scaling in a touch sensitive
device including a touch pad having a sensing range and a display having
a display range according to embodiments disclosed herein may include the
steps of: obtaining a trajectory of touch positions from the touch pad;
setting a first scaling factor; obtaining an acceleration factor
proportional to a speed of motion of a touch; obtaining a deceleration
factor proportional to a measure of an envelope; identifying the location
of a target object in the display; increasing the acceleration factor
when the trajectory overlaps the target object; decreasing the
acceleration factor when the trajectory ceases to overlap the target
object; setting a second scaling factor to the acceleration factor when
the acceleration factor is greater than the deceleration factor; setting
the second scaling factor to the deceleration factor when the first
scaling factor is greater than the deceleration factor and the
acceleration factor is lower than or equal to the deceleration factor;
and updating the trajectory with a new touch position provided by the
touch pad and the second scaling factor.
[0014] According to embodiments disclosed herein a touch sensitive device
coupled to a display, the touch sensitive device having a touch pad and a
controller may include: a processor circuit coupled to receive data from
the touch pad, wherein the processor circuit obtains a touch location
from the data provided by the touch pad; a memory circuit coupled to
receive and store the touch location from the processor circuit and form
a trajectory from a plurality of touch locations; wherein the processor
circuit obtains an instantaneous speed and a moving envelope having a
measure from the trajectory stored in the memory circuit; and the
controller provides a signal to the display to move an indicator to a
position on the display; and the position on the display is obtained by
the processor circuit using the touch location and a scaling factor
computed using the instantaneous speed and the envelope measure.
[0015] These and other embodiments of the present invention are further
described below with reference to the following figures.
BRIEF DESCRIPTION OF THE DRAWINGS
[0016] FIG. 1 illustrates a partial view of a touch sensitive device and a
display according to embodiments disclosed herein.
[0017] FIG. 2A illustrates a flowchart of a method for filtering jitter
motion of a touch sensitive device according to embodiments disclosed
herein.
[0018] FIG. 2B illustrates a jitter-filtered trajectory in a touch
sensitive device according to embodiments disclosed herein.
[0019] FIG. 3 illustrates a direction quantization chart according to
embodiments disclosed herein.
[0020] FIG. 4A illustrates a flowchart of a method for obtaining an
inflection path in a touch sensitive device according to embodiments
disclosed herein.
[0021] FIG. 4B illustrates an inflection path in a touch sensitive device
obtained according to methods disclosed herein.
[0022] FIG. 5A illustrates a flowchart of a method for obtaining a
movement envelope in a touch sensitive device according to embodiments
disclosed herein.
[0023] FIG. 5B illustrates a series of moving envelopes in a touch
sensitive device obtained according to methods disclosed herein.
[0024] FIG. 6A illustrates a motion path and a resulting envelope having a
Dy and a Dx displacement obtained according to embodiments disclosed
herein.
[0025] FIG. 6B illustrates a motion path and a resulting envelope having a
Dy and a Dx displacement obtained according to embodiments disclosed
herein.
[0026] FIG. 6C illustrates a motion path and a resulting envelope having a
Dy and a Dx displacement obtained according to embodiments disclosed
herein.
[0027] FIG. 7A illustrates a flowchart of a method for obtaining a scaling
factor in a touch sensitive device using two terms, according to
embodiments disclosed herein.
[0028] FIG. 7B illustrates a flowchart of a method for obtaining a scaling
factor in a touch sensitive device using two terms, according to
embodiments disclosed herein.
[0029] FIG. 8 illustrates a flowchart of a method for obtaining a
direction-sensitive scaling factor in a touch sensitive device according
to embodiments disclosed herein.
[0030] FIG. 9A illustrates a flowchart of a method for obtaining a
user-sensitive scaling factor having long range performance (LRP),
according to embodiments disclosed herein.
[0031] FIG. 9B illustrates a flowchart of a method for obtaining a
user-sensitive scaling factor having a short range performance (SRP),
according to embodiments disclosed herein.
[0032] FIG. 9C illustrates a flowchart of a method for obtaining scaling
factors balancing an LRP and an SRP, according to some embodiments
disclosed herein.
[0033] FIG. 10 illustrates a display coupled to a touch sensitive device
using a target-aware scaling method, according to embodiments disclosed
herein.
[0034] FIG. 11A illustrates a flow chart of a method for extending a
boundary in a touch sensitive device according to embodiments disclosed
herein.
[0035] FIG. 11B illustrates a flow chart of a method for extending a
boundary in a touch sensitive device according to embodiments disclosed
herein.
[0036] FIG. 11C illustrates a display coupled to a touch sensitive device
using a boundary extension method according to embodiments disclosed
herein.
[0037] FIG. 12A illustrates an apparent position of a finger in a touch
pad coupled to a touch sensitive device configured to avoid an edge
rollback, according to embodiments disclosed herein.
[0038] FIG. 12B illustrates an apparent position of a finger in a touch
pad coupled to a touch sensitive device configured to avoid an edge
rollback, according to embodiments disclosed herein.
[0039] FIG. 12C illustrates an apparent trajectory of a finger in a touch
pad coupled to a touch sensitive device configured to avoid an edge
rollback, according to embodiments disclosed herein.
[0040] FIG. 12D illustrates apparent trajectories of a finger in a touch
pad coupled to a touch sensitive device configured to avoid an edge
rollback, according to embodiments disclosed herein.
[0041] FIG. 13A illustrates a partial side view of a touch sensitive
device configured to avoid an "untouch jump" according to embodiments
disclosed herein.
[0042] FIG. 13B illustrates a partial side view of a touch sensitive
device configured to avoid an "untouch jump" according to embodiments
disclosed herein.
[0043] FIG. 13C illustrates a partial side view of a touch sensitive
device configured to avoid an "untouch jump" according to embodiments
disclosed herein.
[0044] FIG. 14 illustrates a flowchart of a method for avoiding an
"untouch jump" in a touch sensitive device according to embodiments
disclosed herein.
[0045] In the figures, elements having the same designation have the same
or similar functions.
DETAILED DESCRIPTION
[0046] Embodiments disclosed herein relate to the use of a touch sensitive
device where the touch position on a touch pad is translated to a
corresponding position in another domain, such as a screen display. In
some embodiments, the two domains (touch pad/screen display) may have
significantly different resolution or size. In such cases, a one-to-one
mapping between the touch sensitive device and the display is not
practical. An example of such configuration may be a computer touchpad,
where a display cursor (and, by implication, the point at which some
action may be applied by the user) is moved by sliding a finger on a
touch pad. The pad may be smaller and of lower resolution than the
display. If the pad is mapped to the entire display, the user may lose
precision to place the cursor on the display screen. If the pad is mapped
for finer resolution, the user may not reach some portions of the
display.
[0047] In some embodiments, an input touch pad may afford the same or
higher resolution than the display. However, if the touch pad is small,
such as a differential capacitive positioning device, the user may be
unable to precisely position the cursor on the display due to finger
jitter or electronic noise. In this case, the primary problem to be
solved is position jitter resulting from involuntary operator movement,
random environmental capacitance, ground changes, and electrical noise in
the capacitance measuring circuitry. Some embodiments consistent with the
disclosure herein may include a touch screen device, where the mapping
from the touch sensitive device to the display is one to one. This may be
referred to as a 100% scaling.
[0048] Sources of jitter are common in touch sensitive devices; however,
in small size and high resolution devices according to embodiments
disclosed herein, jitter may be substantially amplified on the display.
In high resolution devices, a touch pad position may be mapped to the
entire display without experiencing the quantization error of a lower
resolution touch pad. However, the jitter from such a mapping may
substantially reduce positioning accuracy, regardless of resolution.
[0049] FIG. 1 illustrates a partial view of touch sensitive device 100 and
display 150, according to embodiments disclosed herein. Touch sensitive
device 100 may include touch pad 101, or the sensor input, which is
coupled to controller 103, including processor 102 and memory 104.
Controller 103 provides power to touch pad 101. Touch pad 101 provides a
signal to processor 102 upon being touched by finger 170 from a user.
Relative dimensions in FIG. 1 are for schematic clarity only and do not
necessarily reflect actual sizes. In some embodiments, touch pad 101
includes a capacitive sensing circuit. Display 150 may be coupled to
touch pad 101 through electrical connections in a computer, laptop
device, palm-held electronic device, or some other portable electronic
appliance. Furthermore, display 150 may be coupled to touch pad 101
through a wireless connection operating at a radio frequency (RF), or
through an infrared beam device. In some embodiments, touch pad 101 may
be included as part of display 150.
[0050] A touch by the user may also be executed by other capacitive
devices such as a pen pointer, a stylus, or any other dielectric device
having a narrow tip, according to embodiments consistent with FIG. 1. An
XY coordinate system is shown for ease of reference hereinafter. Choice
of the Y-coordinate along the vertical direction and the X-coordinate
along the horizontal direction is arbitrary and may vary from one
embodiment to another. Processor 102 converts the signal provided by
touch pad 101 into a touch position having (X, Y) coordinates.
Furthermore, processor 102 may also perform mathematical operations to
convert the (X, Y) coordinates from a touch in touch pad 101 into (X',
Y') coordinates of cursor 151 on display 150.
[0051] Display 150 may be a computer display, such as the liquid crystal
display (LCD) on a laptop computer, or the screen in a smart phone or any
other handheld device such as a cell phone, a camera, a PDA or a tablet.
The (X', Y') coordinates on display 150 may represent the position of a
cursor, or a cross-hair, or a visual mark indicating the user a specific
location on display 150. Thus, according to embodiments consistent with
FIG. 1, touch sensitive device 100 may translate finger trajectory 110 on
touch pad 101 onto display trajectory 120 on display 150. While
trajectory 110 starts at point A and ends at point B on touch pad 101,
trajectory 120 starts at point A' and ends at point B' on display 150.
The mathematical transformation from point (X, Y) into point (X', Y') is
performed by processor 102 using a set of parameters including scaling
factor 105 (F), and offset vector 106 (O). Scaling factor 105 and offset
vector 106 may be stored in memory 104, and modified by processor 102.
Scaling factor 105 is a number that relates displacements X and Y in
touch pad 101 to displacements X' and Y' on display 150. Offset vector
106 includes the X'(Ox) and Y'(Oy) coordinates of an offset, position on
display 150. For example, an offset position may be the location of
cursor 151 after a prior touch has been lifted from touch pad 101. Thus,
according to some embodiments, processor 102 may perform the following
mathematical operation on coordinates X, Y to obtain coordinates X', Y':
X'=Ox+FX (1.1)
Y'=Oy+FY (1.2)
[0052] The choice of the origin in coordinate system XY is arbitrary, and
may be adjusted by processor 102 through offset 106. Equations (1.1) and
(1.2) are illustrative only, some embodiments disclosed herein may use a
different combination of operations with scaling factor 105 and offset
106 in order to obtain display position (X', Y') from touch pad position
(X, Y).
[0053] The XY dimensions of touch pad 101, are S.sub.1(111)-S.sub.2(112)
respectively. Likewise, the XY dimensions of display 150 are
D.sub.1(161)-D.sub.2 (162), respectively. Consistent with some
embodiments, display 150 may have a 100% scaling factor, such that the
entire area S.sub.1.times.S.sub.2 of pad 101 is mapped onto the entire
area D.sub.1.times.D.sub.2 of display 150. For example, in some
embodiments touch pad 101 having S.sub.1(111)=10 mm and S.sub.2(112)=10
mm may be fully mapped to display 150 having D.sub.1(161)=400 mm and
D.sub.2(162)=400 mm. In such conditions, the user can position the cursor
anywhere in the display by touching the corresponding point on the input
pad. This affords instantaneous positioning with little precision.
Display 150 may include 2000 addressable points (pixels) in both
directions. If touch pad 101 has equivalent resolution, moving cursor 151
to a specific pixel on display 150 corresponds to finger 170 having a
movement resolution of 5 micrometers. Such resolution is substantially
less than typical finger jitter.
[0054] According to embodiments consistent with the present disclosure
touch sensitive device 100 may include a small touch pad 101 and a larger
display 150, mapping pad 101 to a portion of display 150 with a small F
105. In this case processor 102 implements a variable offset 106 in order
for the user to reach all positions in display 150 from pad 101. For
example, pad 101 may include an area S.sub.1.times.S.sub.2=10.times.10
mm, with cursor 151 starting in the middle of display 150, and finger 170
touching the upper left corner of pad 101. If finger 170 slides
diagonally to the lower right corner of pad 101, the cursor may move the
same real distance (approximately 14 mm) to the lower right corner of the
mapped area if scaling factor 105(F) is one (1). If finger 170 is lifted
from pad 101 to touch down again in the upper left corner of pad 101,
cursor 151 does not move on display 150. Instead, offset vector 106 (O)
is increased accordingly to produce a smooth slide of cursor 151 on
display 150. Using offset vector 106 (O) enables further movement in the
same direction, when a small scaling factor 105 is used.
[0055] However, when small F 105 is used, to traverse the full diagonal in
display 150 finger 170 would have to be repositioned and moved 40 times
in the configuration discussed above. In embodiments disclosed herein the
scaling factor 105 and the offset vector 106 in a controller for a touch
pad are adaptively adjusted in response to apparent user intent. This
enables rapid movement of cursor 151 across a large display area and
precise, jitter-free movement when the user indicates an intention for
it. Scaling changes are made transparent to the user by simultaneously
changing offset vector 106 to retain the current cursor position at every
lift-off motion by finger 170.
[0056] User intent may be detected by sub-perceptive touch gestures.
Sub-perceptive touch gestures are natural and consistent patterns of
touch movement requiring no user training. Unlike perceptive gestures,
sub-perceptive gestures may not correspond to specific application
functions. A two-finger "pinch" is a perceptive gesture that may be
mapped to image size control. Sub-perceptive gestures have a broader
character. Therefore, a larger set of motion characteristics may need to
be determined in order to identify user intent from sub-perceptive
gestures. In order to have an accurate assessment of sub-perceptive
gestures, controller 103 filters touch positions provided by processor
102 from pad 101, to remove motion jitter introduced by finger 170. A
method to achieve jitter motion filtering is described in detail in
relation to FIG. 2A, below.
[0057] FIG. 2A illustrates a flowchart of method 200 for filtering jitter
motion of touch sensitive device 100 according to embodiments disclosed
herein. The steps shown in FIG. 2A may be performed in controller 103 by
processor 102, coupled to memory 104 and receiving data from touch pad
101, according to embodiments consistent with the present disclosure. In
step 205, X and Y accumulators, and a counter (count) are initialized to
a predetermined value. In some embodiments, this predetermined value may
be the same for X, Y accumulators and counter, and equal to zero (0). In
210 `count` is incremented by one (1). In 215, a sample position is
obtained in processor 102 by data provided from touch pad 101. The data
provided by touch pad 101 may be electronic data representing a
capacitance value, or a change in a capacitance value, according to some
embodiments. In some embodiments, the data provided by touch pad 101 may
be electronic data representing the power of an optical signal, or the
power difference between two or more optical signals. In general, the
specific nature of data provided by touch pad 101 may vary according to
different applications of embodiments consistent with the present
disclosure. Obtaining a sample position from data provided by touch pad
101 may include performing mathematical operations with the data in
processor 102. The specific nature of the mathematical operations
performed by processor 102 in order to obtain a touch position (X,Y) in
step 215 vary according to the application of embodiments disclosed
herein. The geometry of touch pad 101 and the nature of the signal
provided, whether a capacitive value, an optical value, or else may also
determine the specific nature of the mathematical operations by processor
102 to obtain (X,Y).
[0058] In step 220, the X accumulator is incremented by the value obtained
in step 215 for the X-coordinate of touch position (X,Y). Likewise, in
step 225 the Y accumulator is incremented by the value obtained in step
215 for the Y-coordinate of touch position (X,Y). The value of `count` is
verified in step 230. If `count`=1, then the sample position, which is
the first point (X.sub.1, Y.sub.1) obtained in step 215, is stored in
memory 104 and method 200 repeats from step 210.
[0059] If `count` is different from one (1), it is then compared to a
predetermined maximum count `max` in step 235. The value of `max` may be
determined by user history, and also by considerations such as the
responsiveness of touch sensitive device 100. A larger value of `max`
allows for a more accurate determination of user intent to move beyond a
jitter range, but may also slow down the responsiveness of touch
sensitive device 100.
[0060] If `count` is less than `max,` in step 240 the distance `D` from
the current sample point (X,Y) to point (X.sub.1,Y.sub.1) is obtained. In
step 245, the value of D is compared to jitter range `JR.` The value of
JR is a predetermined value that represents the range of jitter for a
user. Jitter range JR may be the displacement of finger 170 allowed by
touch sensitive device 100 while assuming the user does not intend to
actually move cursor 151 on display 150. In some embodiments, JR may also
represent inadvertent departures from a desired trajectory of motion by
the user. The value JR is stored in memory 104 and may be modified by
processor 102 according to the history of the user movements. In some
embodiments consistent with FIG. 2A, the value of JR may vary from one
user to another, and memory 104 may keep a registry of JR values for each
user. If step 245 determines that D is less than or equal to JR, then
point (X,Y) is not used to move cursor 151 on display 150. Rather, steps
205 through 235 are repeated as described above. Note that even if cursor
151 is not moved on display 150 the values of X and Y are still collected
in X-accumulator and Y-accumulator, respectively.
[0061] If step 245 determines that D is greater than JR, then the filter
output (Fx, Fy) is updated to sample position (X,Y) in step 250. Thus,
processor 102 uses position (X,Y) to place cursor 151 in position (X',Y')
of display 150. For example, in embodiments consistent with FIG. 2A,
processor 102 may use (X,Y) in formulas such as Eqs. (1.1) and (1.2)
above to obtain (X',Y'). Once the filter output is completed, steps 205
through 235 are repeated, as described above.
[0062] If `count` is greater than or equal to `max` according to step 235,
then the filter output is determined to be (Fx, Fy) in step 255. The
x-coordinate of the filter output, Fx, is the value stored in
X-accumulator divided by the `count` value. Likewise, the y-coordinate of
the filter output, Fy, is the value stored in Y-accumulator divided by
the `count` value. Thus, even if touch position (X,Y) remains within
jitter range from the initial position (X.sub.1,Y.sub.1), the position
(X,Y) may be used by processor 102 to calculate a cursor position
(X',Y'). This may happen if finger 170 lingers near (X.sub.1,Y.sub.1) for
long enough time to accrue a `count` value greater than or equal to
`max.`
[0063] FIG. 2B illustrates jitter-filtered trajectory 110 in touch
sensitive device 100 according to embodiments disclosed herein. Raw
trajectory 260 includes initial point 261 and initial point 262. Also
included in trajectory 260 are sample points 265-1 through 265-5, used by
processor 102 in a jitter filter as described above in relation to FIG.
2A. Filtered trajectory 110 is composed of filtered points 271 and 272
resulting from a filter output (Fx, Fy) according to embodiments
consistent with the present disclosure. Note that, as depicted in FIG.
2B, the precise location of filter output (Fx, Fy), such as points 271
and 272 may not be an actual sample point in the raw trajectory. In some
instances, a filter output (Fx, Fy) may be a sample point, such as when
the distance of a sample point from an initial point is greater than JR.
Note that, according to embodiments consistent with methods described
herein, the distance between points 271 and 272 may be smaller than JR.
In the example shown in FIG. 2B, method 200 has `max`=3, rendering points
271 and 272. Having a trajectory 110 of points in touch pad 101 free of
jitter enables the application of strategies for detecting sub-perceptive
gestures, such as the inflection path strategy.
[0064] Processor 102 produces filtered trajectory 110 including points 271
and 272, and also including time stamps for each of the trajectory
points. Time stamps may be obtained by using a clock circuit included in
processor 102, according to some embodiments. Trajectory 110 and the time
stamps for each point in the trajectory are stored in memory 104. Time
stamps may be used to calculate instantaneous speed of motion for each
point in trajectory 110 in some embodiments.
[0065] Sub-perceptive gestures may be identified by touch position
history. Touch position history may be stored by controller 103 in memory
104. Some embodiments consistent with the present disclosure may include
at least three levels of touch position history. A more immediate level
records the instantaneous direction and speed of movement. A second level
may store recent direction, speed, and distance for a relatively low
number of positions, for example 5 or 6. A third level may include
overall movement pattern using the entire touch position history during a
given period of time. Instantaneous direction and speed are based on
current and previous positions, as obtained from the output of a jitter
filter procedure as illustrated in FIG. 2A. Recent direction and speed
are based on current position and the most recent major inflection point.
Inflection point is where the user deliberately changes direction of
touch movement. According to some embodiments, an inflection includes a
significant change in direction followed by significant movement without
resuming the previous direction. A new recent direction and speed vector
begins at the inflection. Vector history is one form of movement pattern
that uses an inflection path strategy. The inflection path strategy uses
a direction quantization chart as illustrated in detail in FIG. 3.
[0066] FIG. 3 illustrates direction quantization chart 300 according to
embodiments disclosed herein. Chart 300 includes left-right directions
(L-R) 350 and 310, respectively; and up-down directions (U-D) 330 and
370, respectively. Some embodiments may also include intermediate
directions RU (right-up) 320, LD (left-down) 360, LU (left-up) 340, and
RD (right-down) 380. Each of the selected directions 310, 320, 330, 340,
350, 360, 370, and 380 is centered on a corresponding direction interval
315, 325, 335, 345, 355, 365, 375, and 385, respectively. According to
the chart depicted in FIG. 3, a displacement vector (X,Y) having a
certain direction will be assigned one of the 310 through 380 directions
if it lays within its corresponding direction interval. For example, a
displacement vector pointing within direction interval 325 is assigned
direction RU 320, according to embodiments consistent with the present
disclosure. According to embodiments disclosed herein, an `inflection
point` in a trajectory occurs when the direction assigned to a
displacement vector changes from one sample point to another. Thus, if a
first sample point produces a displacement vector in the R 310 direction
and the next sampling point produces a displacement vector in the RU 320
direction, then the first sample point may be characterized as
`inflection point.`
[0067] The characterization of an `inflection point` is arbitrary, and may
vary for different embodiments according to the application used. For
example, in some embodiments an `inflection point` may be characterized
only by drastic changes in direction, such as R 310 to L 350, or U 330 to
D 370.
[0068] FIG. 4A illustrates a flowchart of method 400 for obtaining an
inflection path in touch sensitive device 100 according to embodiments
disclosed herein. In step 405 a time-ordered stream of touch positions
forming trajectory 110 is collected from jitter filter 200 (cf. FIG. 2A).
Method 400 may be implemented by processor 102 using data stored in
memory 104, according to some embodiments consistent with the present
disclosure. Included with trajectory 110 may also be a list of time
stamps for each of the points in trajectory 110. A starting point for a
current inflection leg is selected from trajectory 110 in step 410,
placing a vector tail at this point. An inflection leg is a set including
vectors having a common tail point in an inflection point, and heads
along points in trajectory 110 selected as in steps 420-425, together
with their time stamps. In step 415 the next point in trajectory 110 is
selected, and its distance `Dt` to the previously selected point, is
calculated. In step 420 the value of Dt is compared to a `deliberate
distance.` The `deliberate distance` in step 420 is a pre-selected value
that represents a measure of an intentional movement by the user.
[0069] If Dt is lower than or equal to the `deliberate distance,` then
step 415 is repeated as described above. Steps 415 and 420 are repeated
until a new point (X,Y) in trajectory 110 is obtained such that Dt is
greater than `deliberate distance.` In step 425, the new point (X,Y) is
selected and vector V whose tail is the starting point in step 410 is
completed with its head in selected point (X,Y). Vector V has a size,
.delta..sub.v, and a direction .theta..sub.v. Also in step 425, a
displacement vector S is obtained as the difference between current
vector V and a prior vector V in the current inflection leg. Vector S has
a displacement size .delta..sub.s(=Dt), a displacement direction
.theta..sub.s and a time lapse .tau..sub.s. Time lapse .tau..sub.s may be
obtained by subtracting the time stamps of the two most recently selected
vector heads, and may be used to determine an instantaneous speed v.sub.s
at the selected point as v.sub.s=.delta..sub.s/.tau..sub.s.
[0070] In step 430 a direction change .DELTA..theta. is calculated using
directions .theta..sub.v and .theta..sub.s. Direction change
.DELTA..theta. is compared to a direction tolerance .theta..sub.tol. If
.DELTA..theta. is smaller or equal to .theta..sub.tol, then vector V is
added to the current inflection leg in step 440, together with the time
stamp of point (X,Y). Method 400 then continues as described in step 415.
If .DELTA..theta. is greater than .theta..sub.tol, then a new inflection
leg is started including point (X,Y) and the last point in the previous
inflection leg, in step 435. Thus, the last point of a given inflection
leg may be the first point of the next inflection leg, according to
embodiments consistent with the methods disclosed herein. The first point
of an inflection leg is called `inflection point` I. The distance
.delta..sub.s between the second point in an inflection leg and the
inflection point is .delta..sub.sI. Having started a new inflection leg,
step 410 is repeated, as detailed above, with inflection point I being
the point collected prior to the new point (X,Y).
[0071] In some embodiments, the value .DELTA..theta. obtained in step 430
is the absolute value of the difference between .theta..sub.v for new
point (X,Y) and the most recent point in the current inflection leg.
Also, .theta..sub.tol may be obtained from a direction interval selected
from any of 315, 325, 335, 345, 355, 365, 375, or 385 (cf. FIG. 3). The
direction interval to determine .theta..sub.tol may be selected so that
all vectors in a given inflection leg belong to the same direction
interval. Some embodiments may include a more lax condition to select
.theta..sub.tol in step 430. For example, .theta..sub.tol may be selected
so that all vectors in a given inflection leg belong to two or more
adjacent direction intervals--e.g. intervals 315, 325 and 385, cf. FIG.
3--. In some embodiments consistent with methods disclosed herein,
.DELTA..theta. may be obtained in step 430 using the direction of vector
.theta..sub.s. For example .DELTA..theta. may be the absolute value of
the difference between .theta..sub.s and one of selected directions 310,
320, 330, 340, 350, 360, 370, and 380.
[0072] FIG. 4B illustrates inflection path 450 in touch sensitive device
100 obtained according to method 400. Inflection path 450 includes
inflection legs 401, 402, and 403. Inflection leg 401 includes points
401-1 through 401-7. Inflection leg 402 includes points 402-1 through
402-6. And inflection leg 403 includes points 403-1 through 403-5.
Inflection legs 401-403 are defined by inflection points I: 401-1, 402-1
(401-7), and 403-1 (402-6). It is clear that a change in direction occurs
at each of the inflection points selected. All of the points in
inflection legs 401, 402, and 403 belong to filtered trajectory 110.
However, in some embodiments consistent with methods disclosed herein
some of the points in filtered trajectory 110 may not be included in an
inflection leg.
[0073] FIGS. 4A and 4B illustrate methods and steps to produce a vector
history of filtered trajectory 110 across touch pad 101 such as
inflection path 450. Some embodiments disclosed herein may also use
envelopes of filtered trajectory 110, as described in detail in FIGS.
5A-6C below.
[0074] FIG. 5A illustrates a flowchart of method 500 for obtaining a
movement envelope in touch sensitive device 100 according to embodiments
disclosed herein. In step 505, a number N of points is collected from
filtered trajectory 110, and placed in a circular buffer. The circular
buffer may be included in memory 104, inside controller 103 (cf. FIG. 1).
The value N is an integer number and may be selected according to the
desired accuracy for interpreting sub-perceptive gestures and speed of
response of touch sensitive device 100. A higher value of N may result in
more accurate interpretation of user intent, but a slower response. The N
points (X.sub.1, Y.sub.1) . . . (X.sub.N, Y.sub.N) are ordered according
to their time stamps in trajectory 110. In embodiments consistent with
the present disclosure, a moving envelope of the trajectory includes a
circumscribing polygon for the plurality of points, N, collected in step
505. A moving envelope obtained using method 500 includes the N points
collected in step 505 in the interior of the circumscribing polygon.
Thus, a polygon having any number of sides may be used to obtain a moving
envelope in method 500. In some embodiments the circumscribing polygon is
a convex geometric object having any number of sides, each side having
arbitrary lengths. A convex geometrical object is one that contains in
its interior all points included in a straight line joining any two
points in the interior of the object. According to some embodiments
consistent with the present disclosure, the polygon used for a moving
envelope may be a rectangle.
[0075] In step 510 a set of envelope characteristics is initialized using
the first point (X.sub.1, Y.sub.1) in the buffer. A set of envelope
characteristics may include Xmin, Xmax, Ymin, and Ymax values. Thus, step
510 may include the following operations. Xmin=Xmax=X.sub.1; and
Ymin=Ymax=Y.sub.1.
[0076] In step 515 the next point in the circular buffer is selected, say
(Xi,Yi). Step 520 evaluates if Xi is less than Xmin so that if it is,
then step 525 reassigns Xmin=Xi and skips up to step 540. If Xi is
greater than or equal to Xmin, then step 530 compares Xi to Xmax. If Xi
is larger than Xmax then step 535 reassigns Xmax=Xi. If Xi is smaller
than or equal to Xmax, then method 500 proceeds to step 540. Step 540
evaluates if Yi is less than Ymin so that if it is, then step 545
reassigns Ymin=Yi and skips up to step 560. If Yi is greater than or
equal to Ymin, then step 550 compares Yi to Ymax. If Yi is larger than
Ymax then step 555 reassigns Ymax=Yi. If Yi is smaller than or equal to
Ymax, then method 500 proceeds to step 560. Step 560 verifies if all N
points in the circular buffer have been processed according to steps 520
through 555. If not, then steps 515 through 555 are repeated until the
last point in the buffer is reached. In step 565 the envelope dimensions
are calculated: the X dimension of the envelope is Dx=Xmax-Xmin, and the
Y dimension of the envelope is Dy=Ymax-Ymin. If the last point in the
circular buffer is the last point in trajectory 110 as determined in step
570, then movement envelope procedure 500 is stopped in step 575. If step
575 determines that more points remain in trajectory 110, then a new
point is added at the end of the circular buffer and the first point in
the old buffer is deleted, in step 580. Then, method 500 is repeated from
step 510 until all desired points in trajectory 110 have been included in
at least one envelope.
[0077] FIG. 5B. Illustrates a series of movement envelopes in touch
sensitive device 100 obtained according to methods disclosed herein. For
example, FIG. 5B may be an embodiment of a method such as 500, described
in detail in relation to FIG. 5A, and having a circular buffer of size
N=4. Points 585-1 through 585-8 may belong to trajectory 110 retrieved
from jitter filtering method 200, according to some embodiments. Points
585-1 through 585-8 are included as in method 500, resulting in envelopes
590-A, -B, -C, -D, and -E. Each envelope has x-dimension Dx-A, -B, -C,
-D, and -E, respectively, and y-dimension Dy-A, -B, -C, -D, and -E.
Dimensions Dx and Dy may be used by processor 102 to provide measure `L`
for each envelope. For example, L may be the sum of the Dx and Dy
dimensions, such as L.sub.A=Dx-A+Dy-A. The use of an envelope measure L
will be described in further detail with reference to FIGS. 6A-C, below.
Envelopes 600, 610 and 620 in FIGS. 6A-C may be obtained by a method
including the steps in method 500 as described in detail above.
[0078] FIG. 6A illustrates a motion path and envelope 600 having
Dy.sub.600=0 and a Dx.sub.600=12, according to embodiments disclosed
herein. Points 600-1 through 600-7 in FIG. 6A may belong to trajectory
110 resulting from jitter filter 200, according to some embodiments.
Vectors 601-606 represent displacements S between each of the points in
trajectory 110 forming envelope 600. Displacement vector S was described
above in the context of inflection path 450 (cf. FIG. 4B). In some
embodiments, vector S may include a displacement vector between
successive points in filtered trajectory 110 in the context of a moving
envelope calculation. The resulting envelope measure is
L.sub.600=12+0=12. These measure and envelope dimensions indicate a
consistent path with all vectors following an R 310 direction (cf. FIG.
3).
[0079] FIG. 6B illustrates a motion path and envelope 610 having
Dy.sub.610=6 and a Dx.sub.610=12, according to embodiments disclosed
herein. Points 610-1 through 610-7 in FIG. 6B may belong to trajectory
110 resulting from jitter filter 200, according to some embodiments.
Vectors 611-616 represent displacements between each of the points in
trajectory 110 forming envelope 610. Note that according to the
embodiment illustrated in FIG. 6B, vectors 611-616 have the same lengths
as vectors 601-606 in FIG. 6A. However, in the embodiments illustrated in
FIG. 6B the direction of vectors 611-616 is different from that of FIG.
6A, resulting in different Dx and Dy dimensions for envelope 610. The
resulting measure is L.sub.610=12+6=18.
[0080] FIG. 6C illustrates a motion path and envelope 620 having
Dy.sub.620=3 and a Dx.sub.620=5, according to embodiments disclosed
herein. Points 620-1 through 620-7 in FIG. 6C may belong to trajectory
110 resulting from jitter filter 200, according to some embodiments.
Vectors 621-626 represent displacements between each of the points in
trajectory 110 forming envelope 620. Note that according to the
embodiment illustrated in FIG. 6C, vectors 621-626 have the same lengths
as vectors 601-606 in FIG. 6A and vectors 611-616 in FIG. 6B. However, in
the embodiment illustrated in FIG. 6C the direction of vectors 621-626 is
different from that of FIGS. 6A-B, resulting in different Dx and Dy
dimensions for envelope 620. The resulting measure is L.sub.620=5+3=8.
FIG. 6C may be a sub-perceptive gesture indicating a `hunting` intention
by the user. Starting at the left, the user is trying to move to endpoint
620-7 but overshoots, moving to a position that is higher and to the
right of the target (620-4). The user then moves down to reach point
620-7 but goes too far and overshoots again (620-5). The user moves
leftward (625), aligning with the target (620-6), and then upward (626),
finally hitting the target 620-7. The resulting path produces envelope
620 having less than 2/3 the size of envelope 600 (cf. FIG. 6A) and 610
(cf. FIG. 6B), using the same set of vector lengths 621-626. In such case
a small value of L may indicate a `hunting` action by the user.
[0081] According to embodiments disclosed herein, it may be desirable to
provide a sensing configuration for touch sensor 100 that combines
inflection path method 400 and moving envelope method 500. For example,
some embodiments may use the instantaneous speed measurement v.sub.s in
method 400 to increase scaling factor F 105. In such embodiments,
inadvertently slower movements within a full stroke of finger 170 inherit
F 105 from a faster predecessor stroke. Some embodiments may also include
envelope measures L obtained consistent with method 500 to reduce scaling
factor F 105. Such a configuration will be described in more detail in
relation to FIG. 7A, below.
[0082] FIG. 7A illustrates a flowchart of method 700 for obtaining a
scaling factor in touch sensitive device 100 using two terms, according
to embodiments disclosed herein. A first term may be the instantaneous
speed v.sub.s in an inflection path (cf. FIG. 4A). A second term may be a
measure L from a moving envelope calculation (cf. FIG. 5A). In step 704 a
first scaling factor F 105 is obtained. In some embodiments, a first
scaling factor F 105 may be a scaling factor mapping area
S.sub.1.times.S.sub.2 in touch pad 101 to approximately 50% of area
D.sub.1.times.D.sub.2 in display 150. Some embodiments may have a first
factor F 105 mapping area S.sub.1.times.S.sub.2 in pad 101 to
approximately 100% of area D.sub.1.times.D.sub.2 of display 150. Further,
some embodiments may use a first factor F 105 equal to the last value
stored in memory 104. In step 705, a position sample from a jitter
filtering method is obtained by processor 102. A jitter filtering method
may be as method 200 (cf. FIG. 2A) performed by processor 102. Position
samples resulting from the jitter filter method may be stored in memory
104 and provided to processor 102 in step 705.
[0083] In step 710 displacement vector S is obtained using the new
position sample and the previous position sample provided by memory 104
(cf. step 425 in FIG. 4A). Step 710 may also include the calculation of a
time lapse .tau..sub.s between the new position sample and the previous
position sample, using time stamps associated to each position sample and
stored in memory 104 (cf. step 425 in FIG. 4A). In step 715 the size of
vector S, .delta..sub.s, is determined and compared to a first tolerance
value 715-1. Step 715 may also include comparing .tau..sub.s to a second
tolerance value 715-2. According to some embodiments, if .delta..sub.s is
smaller than the first tolerance and .tau..sub.s is shorter than the
second tolerance value, then the method starts again from step 705 as
described above.
[0084] In some embodiments, once either one of the conditions
.delta..sub.s larger than tolerance 1 and .tau..sub.s larger than
tolerance 2 is satisfied, then an instantaneous speed
v.sub.s=.delta..sub.s/.tau..sub.s is calculated in step 720. In step 722
the circular buffer for envelope calculation is updated with the new
sample point of step 705. With the updated buffer, new envelope
dimensions Dx, Dy, and measure L may be obtained in step 722. Having an
instantaneous speed, acceleration scaling factor AF 701 may be obtained
in step 725. Scaling factor AF 701 is obtained from a product of speed
v.sub.s=.delta..sub.s/.tau..sub.s and a pre-selected acceleration factor.
Scaling factor AF 701 may have values from 1% or less, up to 100%. A
value of 100% for AF 701 maps pad 101 having area S.sub.1.times.S.sub.2
into the entire display 150 having area D.sub.1.times.D.sub.2 (cf. FIG.
1). A value of 1% for AF 701 maps pad 101 having area
S.sub.1.times.S.sub.2 into an area of size
0.01.times.D.sub.1.times.D.sub.2 in display 150. AF 701 may be determined
so that for v.sub.s in a middle range of speeds, a 50% scaling factor is
obtained, for v.sub.s in an upper range of speeds AF 701 is higher than
50%, and for v.sub.s in a lower range of speeds AF 701 lower than 50% is
obtained. A lower, middle, or higher range of speed values may be
established according to prior user history in some embodiments. In some
embodiments, a lower, medium and higher range of speed values may be
determined according to the application of touch sensitive device 100 and
the dimensions of touch pad 101.
[0085] Having envelope dimensions, deceleration scaling factor DF 702 may
be obtained in step 727. DF 702 is obtained from the product of envelope
measure L and a pre-selected deceleration factor. Envelope measure L may
be a sum of an x-dimension and a y-dimension of an envelope, L=Dx+Dy, or
simply either one of the X and the Y dimensions in the envelope.
[0086] In step 730, AF 701 is compared to DF 702. If AF 701 is greater
than DF 702 then AF 701 is compared to current scaling factor F 105 in
step 735. If AF 701 is greater than F 105, then a new scaling factor F
105 is selected as AF 701 in step 740 and method 700 is repeated from
step 704. If AF 701 is less than or equal to F 105 then no adjustment is
done to scaling factor F 105 and method 700 is repeated from step 704. If
AF 701 is less than or equal to DF 702 in step 730, then in step 732 DF
702 is compared to F 105. If DF 702 is smaller than F 105, then in step
737 a new scaling factor F 105 is selected as DF 702, and method 700 is
repeated from step 704. If DF 702 is greater than or equal to F 105 then
no adjustment is done to F105 and method 700 is repeated from step 704.
[0087] According to embodiments consistent with methods disclosed herein,
AF 701 increases F 105 and DF 702 decreases F 105. Some embodiments of
method 700 may include different ways to combine acceleration scaling
factor AF 701 and deceleration scaling factor DF 702 to adjust factor F
105. For example, an updated value of F 105 may be obtained as the
average of AF 701 and DF 702. According to some embodiments consistent
with method 700, if AF 701 is less than current F 105, AF 701 is
discarded. Also consistent with method 700, if DF 702 is greater than
current F 105, DF 702 is discarded. Situations may arise where DF 702 is
less than F 105 and AF 701 is greater than F 105 at the same time. This
will be described in detail with reference to FIG. 7B, as follows.
[0088] FIG. 7B illustrates a flowchart of method 750 for obtaining factor
F 105 in touch sensitive device 100 using two terms, AF 701 and DF 702,
according to some embodiments. Steps 704 through 727 are described in
detail in reference to method 700 above (cf. FIG. 7A). In step 745, DF
702 is compared to the current value of F 105. If DF 702 is greater than
or equal to F 105 then in step 755 AF 701 is compared to current F 105.
If AF 701 is less than or equal to F 105, then the value of F 105 is
unaltered and method 750 is repeated from step 704. If AF 701 is greater
than F 105 in step 755 then F 105 is updated to AF 701 in step 775, and
method 750 is repeated from step 704. If DF 702 is smaller than F 105 in
step 745, then AF 701 is compared to F 105 in step 760. If AF 701 is less
than or equal to F 105 in step 760, then F 105 is updated to be equal to
DF 702 in step 770 and method 750 is repeated from step 704. If AF 701 is
greater than F 105 in step 760, then a situation arises where DF 702 is
less than F 105 and AF 701 is greater than F 105. Such situation may
indicate an unstable finger 170 involuntarily increasing instantaneous
speed. Thus, F 105 is updated with a weighted average of AF 701 in step
765, and method 750 is repeated from step 704.
[0089] According to embodiments of method 750 consistent with FIG. 7B, a
weighted average of AF 701 and DF 702 in step 765 may be in proportion to
their difference from F 105. For example, in some embodiments, scaling F
105 may be updated in step 765 as follows:
new_F = old_F - AF old_F - AF + old_F - DF
.times. AF + old_F - DF old_F - AF + old_F - DF
.times. DF ( 2 ) ##EQU00001##
[0090] Where `new_F` is the updated value of F 105 and `old_F` is the
previous value of F 105.
[0091] In some embodiments, the weighting average of step 765 may be
determined by sub-perceptive gestures in the movement history registered
in memory 104, as follows. The average length of vectors S, or their
average speed v.sub.s=.delta..sub.s/.tau..sub.s may be used to increase
the weight of AF 701 in step 765. Likewise, a history of reducing
envelope measure L=Dx+Dy, may increase the weight of DF 702 in step 765.
In some embodiments, step 765 may be performed over a relatively long
time period in order to collect an accurate weighted average of AF 701
and DF 702. For example, if recent events indicate that the current user
tends to spend an unusual amount of time hunting, weighting can be
adjusted in favor of DF 702. In some embodiments, it may be desirable to
adjust the scaling factor F 105 according to the direction of motion of
trajectory 110. Further, some embodiments may implement a scaling factor
that is different in the X and Y directions of motion across pad 101.
This will be discussed in detail in relation to FIG. 8, below.
[0092] FIG. 8 illustrates a flowchart of method 800 for obtaining a
direction-sensitive scaling factor 105 in touch sensitive device 100
according to embodiments disclosed herein. In step 805 `new scale` factor
NS 801 is selected. In step 810, value XY_fac 802 is selected. In some
embodiments, a full range of XY_fac 802 values may vary from -16 to +16.
In such embodiments the constant XY_fac 802 may be 16. In step 815 a new
envelope of motion is obtained. According to embodiments consistent with
method 800, a new point from jitter filter method 200 may be obtained in
step 815 and added to a circular buffer. Thus, step 815 may include
obtaining a new envelope having dimensions Dx, Dy and measure L,
according to method 500. In step 820 dimension Dy is compared to zero
(0). If Dy is equal to zero, then the motion has been consistently in the
R-L direction (cf. FIG. 3), and a value XY_ratio 803 is selected to be
two times XY_fac 802 in step 825:
XY_ratio=2.times.XY.sub.--fac (3)
[0093] If Dy is different from zero in step 820 then Dx is compared to
zero in step 830. If Dx is equal to zero, then the motion has a
consistent trend in the U-D direction (cf. FIG. 3) and XY_ratio 803 is
set to one (1). If Dx is different from zero in step 830, then XY_ratio
803 is obtained from XY_fac in step 840 as follows:
XY_ratio=(Dx/Dy).times.XY_fac (4)
[0094] Once XY_ratio 803 has been determined in either of steps 825, 835,
or 840, then offset 804 is determined in step 845 as:
offset=(XY_ratio-XY.sub.--fac)/4 (5)
[0095] In step 850, X_range 806 is obtained as:
X_range=(XY_fac+offset).times.new_scale (6)
[0096] In step 855, Y_range 807 is obtained as:
Y_range=(XY_fac-offset).times.new_scale (7)
[0097] In step 860, X_range 806 is compared to D.sub.1 161 (cf. FIG. 1).
If X_range 806 is greater than D.sub.1, then X_range 806 is updated to
D.sub.1 in step 860. If X_range 806 is smaller than or equal to D.sub.1
in 860, then Y_range 807 is compared to D.sub.2 in step 870. If Y_range
807 is greater than D.sub.2, then Y_range is updated to D.sub.2 in step
875. In step 880, scaling factors for the x-coordinate (Fx 808) and the
y-coordinate (Fy 809) are updated using the 806 and 807 values. In some
embodiments, values 808 and 809 may be updated from 806 and 807 by using
the following equations in step 880:
new.sub.--Fx=(x_range/D.sub.1).times.old.sub.--Fx; (8.1)
new.sub.--Fy=(y_range/D.sub.2).times.old.sub.--Fy; (8.2)
[0098] Once updated values 808 and 809 are obtained, processor 102 may use
the updated values to obtain display point (X',Y') based on touch pad
point (X,Y) as follows:
X'=Ox+FxX (9.1)
Y'=Oy+FyY (9.2)
[0099] Some embodiments of adaptive scaling disclosed herein provide fast,
long-range movement and comfortable, short-range focus movement. Fast
long-range movement and comfortable short-range movements are
complementary, thus a control process may use a balanced approach. A
two-term procedure such as methods 700 and 750 (cf. FIGS. 7A and 7B,
respectively), may provide a satisfactory compromise between long-range
and short-range movements. In some embodiments factors AF 701 and DF 702
may be complemented with further factors to produce a more accurate
determination of user intent. A long range performance factor (LRP) and a
short range performance factor (SRP) may be calculated according to
methods described in detail in FIGS. 9A and 9B below.
[0100] FIG. 9A illustrates a flowchart of method 900 for obtaining a
user-sensitive scaling factor 105 having long range performance (LRP),
according to embodiments disclosed herein. In step 902 counters
`new_slide` 921 and `repeat_slide` 922 are initialized. In some
embodiments, counters 921 and 922 may be initialized to a value of zero
(0). In some embodiments, either one of counters 921 and 922 may have a
value different from zero (0). In step 905, processor 102 determines the
completion of a slide motion. In some embodiments, step 905 may be
performed by determining that finger 170 no longer makes contact with
touch pad 101. For example, the signal from a sensor in 101 may fall
below a pre-selected touch threshold level, so that processor 102 may
determine a completion of slide motion in step 905. In step 907 the
coarse direction of the slide motion is determined according to chart 300
(cf. FIG. 3). In some embodiments, the coarse direction of step 907 may
be determined by a displacement vector joining the touch `down` (first)
point in the slide motion and the touch `up` (last) point in the slide
motion. In step 910 the coarse direction determined in step 907 (`new
direction`) is compared with a previously stored value in memory 104 or
`old direction` 901. If `new direction` is different from old direction
901, then new_slide counter 902 is incremented by one (1) in step 915,
and value 901 is updated to `new direction` determined in step 907. If
`new direction` is not different from value 901, then value 901 is kept
`as is` and repeat_slide counter 903 is incremented by one (1) in step
912.
[0101] In step 920 a long range performance value (LRP) 908 is updated. In
some embodiments, step 920 may include performing the following operation
in processor 102:
LRP=repeat_slide/(new_slide+repeat_slide) (10)
[0102] With an updated value LRP 908, method 900 may be repeated from step
902. More generally, LRP 908 measures a percentage of slide motions that
repeat the previous slide direction. A lower LRP 908 value may indicate
better response of touch sensitive device 100 to user intent of moving
cursor 151 through long distances.
[0103] FIG. 9B illustrates a flowchart of method 920 for obtaining a
user-sensitive scaling factor having a short range performance (SRP),
according to embodiments disclosed herein. In step 921 a value for
`reversal` counter is initialized. In some embodiments, `reversal` is set
to zero (0) in step 921. In step 922 touch point (X,Y) is obtained and it
is determined if it is an inflection point I. In some embodiments,
processor 102 obtains point (X,Y) after performing jitter filtering
method 200 (cf. FIG. 2A). In step 922 inflection point I includes a
displacement S to a most recent point (X,Y), S having a direction and a
length .delta..sub.sI. In step 923 length .delta..sub.sI is compared to
tolerance 923t. If .delta..sub.sI is less than or equal to 923t then in
step 925 .delta..sub.sI is compared to minimum value 925min. If
.delta..sub.sI is greater than 925min in step 925 then the direction of S
(`new direction`) is compared to the old direction in step 927, according
to chart 300 (cf. FIG. 3). The old direction of vector S is a value
stored in memory 104 according to chart 300 corresponding to a previous
value of S. If vector direction is reversed; that is, old direction is
opposite to `new direction,` then reversal counter is incremented by one
(1) in step 930 and method 920 is repeated from step 922. If `new
direction` is not reversed according to step 927, then method 920 is
repeated from step 922 leaving reversal counter unchanged. If
.delta..sub.sI is less than or equal to 925min in step 925, then method
920 is repeated from step 922 leaving reversal counter unchanged.
[0104] If .delta..sub.sI is greater than 923t in step 923 then `new
direction` is compared to `old direction` in step 932. If `new direction`
is not opposite to `old direction` then method 920 is repeated from step
922, leaving reversal counter unchanged. If `new direction` is reversed
in step 932 then status variable 935stat is compared to the value
`hunting` in step 935. If 935stat is `hunting` then SRP value 945 is
updated to reversal counter in step 937, and 935stat is updated to `not
hunting` in step 940. Method 920 is repeated from step 922, leaving
reversal counter unchanged. If 935stat is `not hunting` in 935, then
reversal counter is set to zero (0) in step 942 and 935stat is updated to
`hunting` in step 945. Method 920 is repeated from step 922.
[0105] A higher value of SRP 945 indicates an increased degree of
`hunting` by the user. Thus, lower SRP 945 may indicate better response
of touch sensitive device 100 to user intent to move the cursor short
distances. LRP 908 and SRP 945 have no units or inherent absolute values
according to some embodiments. LRP 908 and SRP 945 may be compared by
subtraction or division, yielding a measure of their balance. A balance
between LRP 908 and SRP 945 may be consistent with user intent. Scaling
coefficients may be continuously adjusted by processor 102 to achieve a
particular balance level. One possible realization of this balance is
described in detail in relation to FIG. 9C below.
[0106] FIG. 9C illustrates a flowchart of method 950 for obtaining scaling
factors balancing LRP 908 and an SRP 945, according to embodiments
disclosed herein. Method 950 may be performed by processor 102 using data
provided by touch pad 101 and stored in memory 104. In step 951,
pre-selected coefficients k1 and k2 are obtained. In step 952 a new point
(X, Y) is obtained. In some embodiments, new point (X, Y) may be the
result of a jitter filter procedure such as method 200 performed by
processor 102 using touch sensing data provided by pad 101. In step 955
balance 971 is computed using k1, k2, LRP 908, and SRP 945. According to
embodiments consistent with method 950, balance 971 may be computed
according to the following formula:
balance=k1LRP-k2SRP (11)
[0107] In step 957, balance 971 is compared to zero (0). If balance 971 is
equal to zero, then no adjustments are performed and method 950 is
repeated from step 952. If balance 971 is found to be less than zero in
step 960, then processor 102 decreases AF 701 and increases DF 702 in
step 962. The consequent decrease in the weighting of acceleration (FIG.
7A step 725) and increase in the weighting of deceleration (FIG. 7A step
727) in the scaling method described by FIG. 7A may decrease the effort
required to achieve a target position. This in turn may result in a
decrease of SRP 945 (indicating better short-range performance)--cf. the
description of method 920 in FIG. 9B--and an increase of LRP 908
(indicating worse long-range performance)--cf. the description of method
900 in FIG. 9A--resulting in balance 971 moving toward zero (cf. Eq. 11).
If balance 971 is found to be greater than zero in step 960, then
processor 102 increases AF 701 and decreases DF 702 in step 965. The
resulting increase in the weighting of acceleration (FIG. 7A step 725)
and decrease in the weighting of deceleration (FIG. 7A step 727) in the
scaling method described by FIG. 7A may decrease the effort required to
move long distances. This in turn may result in a decrease of LRP 908
(indicating better long-range performance)--cf. the description of method
900 in FIG. 9A--and an increase of SRP 945 (indicating worse short-range
performance)--cf. the description of method 920 in FIG. 9B--resulting in
balance 971 moving toward zero (cf. Eq. 11). Method 950 is then repeated
from step 952.
[0108] Adaptive scaling based on touch movement patterns such as
instantaneous speed and envelope consistent with method 950 described
above affords robust control of touch sensitive device 100. Some
embodiments of touch sensitive device 100 may include a controller 103
able to single out special situations commonly arising in the use of
device 100. Some embodiments of controller 103 may include target-aware
scaling described in relation to FIG. 10 below.
[0109] FIG. 10 illustrates display 150 coupled to touch sensitive device
100 and using target-aware scaling method 1000, according to embodiments
disclosed herein. Display 150 may include multiple targets 1001 separated
by inactive screen areas. FIG. 10 illustrates a number of target objects
1001-1 through 1001-9 placed on display 150. Targets 1001 can vary in
size and, therefore, level of touch acuity required for reliable
selection, while inactive areas require very little acuity. Adaptive
scaling based on touch movement patterns reacts to the user's response to
target arrangement. This makes the control of touch sensitive device 100
reactive, typically having a lag time. For example, a lag may appear as
delayed acceleration when the user wants to move quickly and delayed
deceleration causing overshoot and forcing the user to hunt for a target.
If controller 103 is aware of the varying acuity requirements of display
150, it can anticipate user intent and adjust scaling appropriately.
[0110] A slide 1050 on display 150 resulting from filtered trace 110 on
touch pad 101 may have a starting point on target 1001-7 and an ending
point on target 1001-3. Slide 1050 crosses over targets 1001-7, 1001-5,
and 1001-3, and also over inactive portions of display 150 in segments
1010-1 and 1010-2. In such a situation, controller 103 may increase F 105
in portions 1010-1 and 1010-2, and reduce F 105 when slide 1050 is
crossing over, or in the vicinity of targets 1001-7, 1001-5 and 1001-3.
[0111] Consistent with some embodiments, accuracy may be enhanced by
establishing a target preference. Thus, if 1001-3 were a more likely
target than 1001-5 for a given user, F 105 could be kept as in inactive
area 1010-1 and 1010-2 as the cursor travels over 1001-5. Knowledge of
user preference for a given target may be provided to controller 103 by
the application itself, or learned by controller 103 from behavior
history. Accordingly, in some embodiments an unintended increase in F 105
around 1001-5 may induce an increase in `hunting.` This in turn may
reduce SRP 945 (cf. FIG. 9B). This may be corrected in embodiments
consistent with method 950 (cf. FIG. 9C). In some embodiments, controller
103 may learn target preferences by storing in memory 104 areas of
display 150 having slow finger movement, rapid direction reversals
(indicating `hunting`), and high selection rates (by tap or other means).
[0112] Touch movement acuity to comfortably manipulate a particular
display can vary. In some embodiments, coarse positioning (large F 105)
may be used to select images from a thumbnail gallery. In some
embodiments, an application may provide controller 103 with the level of
acuity required, so F 105 is adjusted accordingly for the entire display.
In embodiments consistent with method 700 (cf. FIG. 7A), to decrease
acuity AF 701 may be increased. Also, a decrease in DF 702 may decrease
acuity. Likewise, to increase acuity AF 701 may be decreased; and an
increment in DF 702 may also increase acuity.
[0113] Some embodiments of controller 103 may include boundary extension
detection as a special situation arising in touch sensitive device 100.
This will be described in relation to FIGS. 11A-C.
[0114] FIG. 11A illustrates flow chart of method 1100 for extending a
boundary in a touch sensitive device according to embodiments disclosed
herein. Cursor 151 may move close to a target 1001 on display 150 (cf.
FIG. 10), but finger 170 may be already on the edge of touch pad 101.
Method 1100 may be used in some embodiments to correct a situation where
cursor 151 is close to target 1001 but the current mapping of pad 101 to
display 150 does not include target 1001.
[0115] According to embodiments shown in FIG. 11A a new position sample is
obtained from jitter filter 200 in step 1102. If the finger is not near
the edge of touch pad 101 in step 1105 then a scaling adder is set to
zero (0) in step 1107 and method 1100 is repeated from step 1102. If the
finger is near the edge of touch pad 101 in 1105, then controller 103
verifies if memory 104 has recorded a significant movement away from the
edge, in step 1108. If this is the case, then the scaling adder is set to
zero (0) and method 1100 is repeated from step 1102.
[0116] If no significant movement away from the edge is detected in step
1108 then status variable 1110stat is queried in step 1110. According to
embodiments consistent with method 920, variable 1110stat may be the same
as variable 935stat in step 935 (cf. FIG. 9B). If 1110stat is `hunting`
in step 1110 then the scaling adder is set to zero (0) in step 1112 and
method 1100 is repeated from step 1102. If 1110stat is not `hunting` in
1110 then it is determined that the finger movement intends to approach
the edge of touch pad 101 in step 1115. In step 1117 the scaling adder is
compared to zero (0). If scaling adder is zero (0) in step 1117, then a
value different from zero (0) is assigned to the scaling adder in step
1122 and F 105 is updated by adding the scaling adder to the previous
value of F 105. Method 1100 is then repeated from step 1102. If scaling
adder is different from zero (0) in 1117 then F 105 is left unchanged in
step 1120 and method 1100 is repeated from step 1102.
[0117] FIG. 11B illustrates a flow chart of method 1130 for extending a
boundary in touch sensitive device 100 according to embodiments disclosed
herein. Steps 1102 and 1105 are as described above in relation to method
1100 (cf. FIG. 11A). If finger 170 is near an edge of touch pad 101 in
step 1105, then in step 1131 controller 103 establishes if finger 170 is
moving towards the edge. Controller 103 may perform step 1131 by
obtaining the direction of vector S between two consecutive touch
positions (cf. FIGS. 3 and 4B). If no indication to stop is detected in
step 1135, then controller 103 continues in step 1140 to move cursor 151
on display 150, even if the displacement of finger 170 is smaller than JR
(cf. FIG. 2B). If an indication to stop is detected in step 1135, then
cursor 151 is stopped on display 150 in step 1137 and method 1130 is
repeated from step 1102.
[0118] According to embodiments consistent with methods disclosed herein
an indication to stop in step 1135 may be any movement of finger 170
opposite to the current direction of motion. This may be detected by
comparing the direction of a current displacement vector S with the
direction of a displacement vector S stored in memory 104. In some
embodiments, an indication of user intent to stop in step 1135 may be
obtained by a decrease in touch pressure. This may be provided by
measurement from a pressure transducer coupled to touch pad 101, or by
estimating the touch area in touch pad 101. A reduction of the touch area
may be an indication to stop in step 1135.
[0119] FIG. 11C illustrates display 150 coupled to touch sensitive device
100 using a boundary extension method according to embodiments disclosed
herein. Finger 170 is moving towards the edge of touch pad 101 along
trajectory 110. Trajectory 110 may be a jitter filtered trajectory after
controller 103 applies method 200. Display 150-1 shows a situation where
F 105 is left unchanged and therefore the boundary of touch pad 101
translates into boundary 1160-1. Desired target 1150-1 may remain outside
of translated boundary 1160-1 and translated trajectory 120-1 may not
reach target 1150-1. In embodiments consistent with method 1100 (cf. FIG.
11A) or method 1130 (cf. FIG. 11B) F 105 may be adjusted accordingly so
that target 150-2 is within translated boundary 1160-2 and translated
trajectory 120-2 is able to reach over target 1150-2.
[0120] Some embodiments of touch sensitive device 100 disclosed herein may
implement methods for distinguishing between finger movement and apparent
position changes resulting from sensing device artifacts. Touch sensing
transducers with clearly defined edges, such as optical, reliably
indicate when finger 170 reaches an edge of touch pad 101. Transducers
with lower edge definition, particularly capacitive, may detect edges by
including further analysis in controller 103. For example, in a
differential capacitance device when finger 170 is located between two
opposing electrodes in touch pad 101, each electrode sees approximately
the same capacitance. As finger 170 moves away from the middle, for
example in the L direction (cf. FIG. 3), capacitance in the left
electrode increases while that of the right electrode decreases. This may
continue until finger 170 has moved close to the left edge of the left
electrode. At this point capacitance on the right electrode is
significantly independent of finger motion. Finger 170 may continue to
affect capacitance on the left electrode. A further leftward movement by
finger 170 may decrease the capacitance in the left electrode more
drastically than in the right electrode. Thus, in some situations this
may be interpreted erroneously as a movement in the R direction. This
effect may be referred to as `edge rollback.` Embodiments to resolve edge
rollback situations are described in detail in relation to FIGS. 12A-D,
below.
[0121] FIGS. 12A-D show a physical action of finger 170 moving across
touch pad 101, and a controller representation of the movement. The
controller representation is created within controller 103 as a set of
(X,Y) positions that the controller will convert into (X',Y') positions
for cursor 151 on display 150. The controller representation includes a
set of points (X,Y) that form representation 101' of the geometric
configuration of touch pad 101. According to embodiments consistent with
the present disclosure, touch pad 101 may be a capacitive coupled sensor
including four electrodes. The four electrodes in touch pad 101 may be
arranged symmetrically in U(up), D(down), L(left), and R(right)
configuration. Some embodiments consistent with the present disclosure
may use other type of sensor devices in touch pad 101, such as an
optically coupled sensor, or a pressure transducer. While the detailed
description below may refer to a capacitively coupled electrode,
embodiments included herein are not limited to these specific devices.
[0122] FIG. 12A illustrates apparent position 1201 of finger 170 in touch
pad 101' coupled to touch sensitive device 100 configured to avoid an
edge rollback, according to embodiments disclosed herein. In FIG. 12A
finger 170 is near the center of touch pad 101 and apparent touch
position 1201 accurately represents the location of finger 170 within
touch pad 101'.
[0123] FIG. 12B illustrates apparent position 1202 of finger 170 in touch
pad 101' coupled to touch sensitive device 100 configured to avoid an
edge rollback, according to embodiments disclosed herein. As finger 170
moves close to the left edge of touch pad 101, position 1202 accurately
represents the location of finger 170 on the left edge of touch pad 101'.
[0124] FIG. 12C illustrates apparent trajectory 1203 of finger 170 in
touch pad 101' coupled to a touch sensitive device configured to avoid an
edge rollback, according to embodiments disclosed herein. As finger 170
moves out to the left edge of touch pad 101 (physical action), the touch
position in controller 103 follows apparent trajectory 1203, close to the
central portion of touch pad 101' (controller representation). The
relatively large decrease in left electrode capacitance in touch pad 101
compared to the small decrease in right electrode capacitance makes
apparent trajectory 1203 in controller 103 move toward the center of
touch pad 101'. Trajectory 1203 is an artifact that may not represent the
true movement of finger 170 on pad 101.
[0125] FIG. 12D illustrates apparent trajectories 1204-1 and 1204-2 of
finger 170 in touch pad 101' coupled to touch sensitive device 100
configured to avoid an edge rollback, according to embodiments disclosed
herein. Trajectory 1204-1 traces the raw position path during a `phantom`
move from apparent position 1202 in FIG. 12B to apparent trajectory 1203
in FIG. 12C. Trajectory 1204-1 is raw data with no jitter filter 200
applied to it. In embodiments consistent with edge rollback avoidance
strategy as described in FIGS. 12A-D, the jitter may be used as an
indicator of user intent, as follows. Jitter in 1204-1 is random and
relatively small. Trajectory 1204-1 is consistently rightward while
finger 170 consistently moves leftward (cf. `physical action` in FIG.
12C). Trajectory 1204-2 traces the unfiltered position path resulting
from withdrawing the finger from touch pad 101 after the position shown
in FIG. 12B. This type of motion may occur if the user is not intent on
continuing a slide motion of cursor 151 on display 150. Random
displacements in 1204-2 are substantially larger than normal jitter. The
direction of trajectory 1204-2 is also random and shows no consistent
orientation. If finger 170 remains on the left edge of the left electrode
as in FIG. 12B, raw movement shows small random jitter vectors such as
trajectory 1204-1.
[0126] Thus, a method for controller 103 to differentiate between user
intent to keep a slide motion to the left of touch pad 101 and the
intention to start a different slide motion may be as follows. If touch
position approaches the left edge of 101' as in positions 1201 and 1202,
and then moves consistently rightward with only small random jitter as in
1204-1, this may indicate intent to keep a leftward slide motion. Some
embodiments consistent with the present disclosure may use a pre-selected
value for jitter threshold so that if jitter is less than or equal to the
threshold then the user intent is determined to be `continue motion`
(leftward, according to FIG. 12 C). According to some embodiments, the
jitter threshold may be determined in a calibration procedure, and thus
the value of jitter threshold may vary according to application and user
habits. Furthermore, some embodiments may include other sensor
measurements to complement the above determination. For example, in a
capacitively coupled touch sensor measuring an approximately monotonic
decrease in overall capacitance may indicate finger 170 intent on keeping
a leftward slide motion.
[0127] Some embodiments consistent with the disclosure herein may include
high resolution differential capacitive touch sensors. In such cases, and
also in other embodiments consistent with the present disclosure, phantom
positioning may occur when touching finger 170 is withdrawn (untouch)
from pad 101. This phantom positioning was described in detail with
respect to apparent trajectory 1204-2 in FIG. 12D, above.
[0128] FIG. 13A illustrates a partial side view of touch sensitive device
100 configured to avoid a untouch jump according to embodiments disclosed
herein. FIG. 13A shows a coordinate axis having XZ coordinates to
illustrate vertical displacement Dz 1307 between finger 170 and touch pad
101. Coordinate axis XZ in FIG. 13A is consistent with a 3D
`right-handed` coordinate system XYZ of which the XY portion is shown in
FIG. 1. The choice of coordinate axes XYZ in FIGS. 1 and 13A-C is
non-limiting of the embodiments disclosed herein. The particular
selection of coordinates XYZ disclosed herein is made for ease of
description. As illustrated in FIG. 13A, finger 170 is in full contact
with pad 101 so that Dz 1307 is zero (0). Thus, the capacitive influence
range of finger 170 may encompass the entire device (if the device is
small relative to the finger) or a well-defined area of the device (if
the device is much larger than the finger). Finger 170 has some influence
at all distances, but in some embodiments controller 103 may ignore
capacitance effects below pre-selected threshold dCt. Threshold dCt
determines sensitive region 1305 around finger 170. When finger 170 is in
full contact with pad 101, the capacitance in pad 101 raises above dCt
and controller 103 registers the finger position. In configurations where
finger 170 is not fully in contact with pad 101 but is within region 1305
of pad 101, a touch may still be registered by controller 103, as shown
in FIG. 13B described in detail below.
[0129] FIG. 13B illustrates a partial side view of touch sensitive device
100 configured to avoid an "untouch jump" according to embodiments
disclosed herein. According to embodiments consistent with FIG. 13B
distance Dz 1307 may be different from zero (0). Still, touch pad 101 may
be within region 1305 from finger 170 and controller 103 registers a
touch position for this configuration.
[0130] FIG. 13C illustrates a partial side view of touch sensitive device
100 configured to avoid an "untouch jump" according to embodiments
disclosed herein. When finger 170 is positioned far off pad 101, as shown
in FIG. 13C, region 1305 may not include touch pad 101. When finger 170
is withdrawn from the device from a configuration as in FIG. 13B to a
configuration as in FIG. 13C, the capacitive effect of finger 170
transitions from above dCt to below dCt. During this transition the
capacitive effect of finger 170 may vary randomly and rapidly. If
controller 103 registers this motion, it may produce a phantom movement
such as trajectory 1204-2 (cf. FIG. 12D). The result may be a substantial
position jump at an untouched location. A method for solving this problem
is described in detail in relation to FIG. 14, below.
[0131] FIG. 14 illustrates a flowchart of method 1400 for avoiding an
"untouch jump" in a touch sensitive device according to embodiments
disclosed herein. According to some embodiments, method 1400 may be
performed by controller 103 using processor 102, data provided by touch
pad 101 and stored in memory 104. In step 1405 controller 103 obtains a
new finger position (X,Y) resulting from jitter filtering method 200. In
step 1410 controller 103 calculates the instantaneous speed from the
finger movement. In some embodiments consistent with method 400,
controller 103 performs step 1410 using v.sub.s=.delta..sub.s/.tau..sub.s
(cf. FIG. 4B). In step 1415 controller 103 calculates a moving average of
speeds in prior movement points. In some embodiments controller 103 may
use data stored in memory 104, to perform step 1415. In step 1420,
controller 103 calculates cap_hist 1401 as a moving average of total
capacitance values at different points of filtered trajectory 110.
Controller 103 may perform step 1420 using data stored in memory 104. In
step 1425 controller 103 calculates cap_diff 1402 as the difference
between cap_hist 1401 and the total capacitance of new position (X,Y).
[0132] In step 1430 value 1402 is compared to pre-selected threshold Th
1403. If cap_diff 1402 is less than or equal to Th 1403 then a status
variable is set to `touch` in step 1435, cursor 151 is moved on display
150 in step 1437 and method 1400 is repeated from step 1405. Thus, when a
touch position is provided through jitter filter 200 and there is no
change in capacitance above threshold Th 1403, controller 103 assumes the
touch continues. Note that according to embodiments consistent with
method 1400 and FIG. 13B, controller 103 may register touch positions
even when Dz is different from zero (0). If cap_diff 1402 is greater than
Th 1403 in step 1430, then controller 103 calculates speed_change 1470 in
step 1440. In some embodiments speed_change 1470 may be calculated using
processor 102 to perform the following formula:
speed_change=vs/<vs> (12)
[0133] where vs is the instantaneous speed calculated in step 1410, and
<vs> is the moving average speed calculated in step 1415. In step
1445 speed_change 1470 is compared to pre-selected threshold Th2 1475. If
speed_change 1470 is less than or equal to Th2 1475 then controller 103
moves cursor 151 in step 1437 and method 1400 is repeated from step 1405.
In step 1445 controller 103 determines that even though a capacitance
change has occurred, the instantaneous speed of the movement has not
changed beyond threshold Th2 1475 and assumes that the touch continues.
[0134] If speed_change 1470 in step 1445 is larger than Th2 1475 then a
combination of speed_change 1470 and cap_diff 1402 is compared to Th3
1480 in step 1455. In some embodiments, the combination of speed_change
1470 and cap_diff 1402 may be a sum of the two values. If the combination
of speed_change 1470 and cap_diff 1402 is less than or equal to Th3 1480
then controller moves cursor 151 in step 1437 and method 1400 is repeated
from step 1405. Thus, in step 1455 controller 103 may determine that even
if cap_diff 1402 is larger than Th 1403 and speed_change 1470 is larger
than Th2 1475, a combination of both may not be larger than Th3 1480.
Thus controller 103 may assume that the touch slide continues. If the
combination of cap_diff 1402, and speed_change 1470 is larger than Th3
1480 in step 1455, then controller 103 updates the status variable to
`untouch` in step 1465. Method 1400 is then repeated from step 1405
without moving cursor 151 on display 150.
[0135] According to some embodiments, method 1400 may be beneficial when
touch pad 101 is a small capacitive device coupled to a large display
150. Some embodiments may use method 1400 in full touch screens having
essentially 100% scaling. Some embodiments may use method 1400 in touch
pads which work effectively with low or substantially zero (0)
acceleration. Method 1400 may also be used in embodiments where touch
sensor 101 includes optical devices. In such cases phantom movement
produced by a defocusing effect at untouch may be avoided using method
1400.
[0136] In the figures, elements having the same designation have the same
or similar functions. Embodiments of the invention described above are
exemplary only. One skilled in the art may recognize various alternative
embodiments from those specifically disclosed. Those alternative
embodiments are also intended to be within the scope of this disclosure.
As such, the disclosure is limited only by the following claims.
Prev
Home