CROSS-REFERENCE TO RELATED APPLICATIONSThis application is a continuation of U.S. patent application Ser. No. 12/880,805 filed Sep. 13, 2010, which is a continuation of U.S. patent application Ser. No. 11/253,477 filed on Oct. 18, 2005 (now U.S. Pat. No. 7,817,135), which is a continuation of U.S. patent application Ser. No. 10/382,799, filed Mar. 5, 2003 (now U.S. Pat. No. 7,212,189), which is a continuation of U.S. patent application Ser. No. 09/971,181, filed Oct. 4, 2001, which is a divisional of U.S. Pat. No. 6,587,093, filed Nov. 3, 2000, which claims priority to U.S. Provisional Application Ser. No. 60/163,635, filed Nov. 4, 1999.
BACKGROUND OF THE INVENTIONThis patent discloses a computer mouse implemented partially or wholly using capacitive sensors. Pointing devices are an essential component of modern computers. One common type of pointing device is the mouse. Computer mice have been well known for many years. U.S. Pat. No. 3,541,541 to Engelbart discloses an early mouse implementation using either potentiometers or wheels with conductive patterns to measure the motion. The conductive patterns on these wheels are measured by direct electrical contact. Direct electrical contact to moving objects has many well-known disadvantages, such as increased friction, and wear and corrosion of contacts.
Modern mice follow a plan similar to that of U.S. Pat. No. 4,464,652 to Lapson et al, with a rolling ball mechanically coupled to optical rotary motion encoders. The mouse also includes one or several buttons that operate mechanical switches inside the mouse. Recent mouse designs also feature a wheel for scrolling; U.S. Pat. No. 5,530,455 to Gillick et al discloses a mouse with a scroll wheel mechanically coupled to another optical rotary encoder. Such mechano-optical mice are widely used and well understood, but they do suffer several drawbacks. First, as moving parts they are susceptible to mechanical failure and may need periodic cleaning. Second, they are exposed to dirt, moisture, and other contaminants and environmental effects. Third, as low-cost mechanical devices they may be less sensitive to fine movements than fully electronic devices. Fourth, electromechanical sensors may be more expensive than purely electronic sensors. And fifth, optical sensors draw a significant amount of power due to their use of light emitting diodes.
Another well-known type of mouse measures motion by direct optical sensing of the surface beneath the mouse. U.S. Pat. No. 4,364,035 to Kirsch discloses an optical mouse that worked with patterned surfaces, and U.S. Pat. No. 5,907,152 to Dandiker et al discloses a more sophisticated example that works with natural surfaces. U.S. Pat. No. 5,288,993 to Bidiville et al discloses a pointing device which includes a rotation ball but measures the rotation of the ball by purely optical means. Optical mice eliminate the difficulties associated with moving parts in the motion sensor, but even they must typically use mechanical mouse buttons and a mechanical scroll wheel.
Many alternatives to scroll wheels have been tried. U.S. Pat. No. 5,883,619 to Ho et al discloses a mouse with a four-way scrolling button. U.S. Pat. No. 5,313,229 to Gilligan et al discloses a mouse with a thumb-activated scrolling knob. U.S. Pat. No. 5,122,785 to Cooper discloses a mouse that is squeezed to initiate scrolling. The ScrollPoint Mouse from International Business Machines includes an isometric joystick for scrolling, and the ScrollPad Mouse from Fujitsu includes a resistive touch sensor for scrolling. The proliferation of such devices shows both that there is a need for a good scrolling device for use with mice, and that none of the technologies tried so far are completely satisfactory.
Capacitive touch pads are also well known in the art; U.S. Pat. No. 5,880,411 discloses a touch pad sensor and associated features. Touch pads can simulate the motion detector and buttons of a mouse by measuring finger motion and detecting finger tapping gestures. Touch pads can also be used for scrolling, as disclosed in U.S. Pat. No. 5,943,052. Capacitive touch pads are solid state electronic devices that avoid many of the pitfalls of mechanical sensors. However, many users prefer mice over touch pads for reasons of ergonomics or familiarity.
Capacitive touch sensors for use as switches are well known in the art. For example, U.S. Pat. No. 4,367,385 to Frame discloses a membrane pressure switch that uses capacitance to detect activation. U.S. Pat. No. 5,867,111 to Caldwell et al discloses a capacitive switch that directly detects the capacitance of the user. The circuits of the '411 patent already cited could also be used to implement a capacitive switch. Applications of capacitive switches to mice are relatively rare, but in the paper “Touch-Sensing Input Devices” (ACM CHI '99, pp. 223-230), Hinckley and Sinclair disclose an experimental mouse with capacitive touch sensors to detect the presence of the user's hand on or near various mouse controls.
U.S. Pat. No. 5,805,144 to Scholder et al discloses a mouse with a touch pad sensor embedded in it. However, Scholder only considers resistive and thermal touch sensors, which are less sensitive and less able to be mounted within the plastic enclosure of the mouse than capacitive sensors. Scholder suggests using the touch sensor in lieu of mouse buttons, but does not consider the use of the touch sensor for scrolling.
The purpose of the present invention is to create a device with the familiar form and function of a mouse, wherein some or all of the mechanical functions of the mouse have been replaced by capacitive sensors.
SUMMARYThe present invention is directed toward a capacitive sensing device for effecting a user interface action based on the measured variations of capacitance. In an embodiment, the device includes a touch surface, a capacitive touch sensor coupled to the touch surface and configured to measure finger motion along the touch surface, and a processor in operative communication with the capacitive touch sensor. The processor is configured to: generate a scrolling command in response to finger motion along the touch surface; cease the scrolling command without substantially continuing generating the scrolling command upon the finger lifting from the touch surface when the finger is stationary prior to lifting from the touch surface; and continue generating the scrolling command for a time after the finger lifting from the touch surface to emulate coasting responsive to finger motion prior the finger lifting.
The disclosed device is directed towards a computer mouse. The computer mouse comprises a touch sensor embedded within a surface material of the mouse. The touch sensor is configured to measure motion of a finger along an axis. The touch sensor is configured to operate by capacitive means.
Another embodiment disclosed includes a pointing device. The pointing device comprises a computer mouse configured to generate cursor commands. A touch sensor is coupled to the computer mouse. The touch sensor is configured for measuring motion of a finger along an axis. The touch sensor is configured for operating by capacitive means. A processor is in operative communication with the touch sensor. The processor is configured to generate a scrolling command in response to the motion of the finger along the axis. The processor is configured to continue generating the scrolling command responsive to the finger lifting from the touch sensor.
Another embodiment disclosed includes a touch input system. The touch input system comprises a capacitive touch sensor configured for measuring motion of a finger along an axis. A processor is in operative communication with the capacitive touch sensor. The processor is configured to generate quadrature signals compatible with those from an optical rotary motion encoder in response to the motion of the finger along the axis.
Yet another embodiment disclosed includes a one-axis touch sensor configured for sensing an object along a single axis. The one-axis touch sensor is configured to generate a scrolling signal responsive to sensing motion of the object touching the one-axis touch sensor.
Still another embodiment disclosed includes a one-axis touch sensor comprising a sensor configured to sense along a single axis. The sensor is configured to generate a quadrature signal responsive to an object touching the sensor. The quadrature signal including characteristics of signals being of the type produced by a rotary encoder.
Still another embodiment disclosed includes a one-axis touch sensor comprising a sensor configured to sense a finger along a single axis of the one-axis touch sensor. A processor is in operative communication with the sensor. The sensor is configured to transmit to the processor one of a touch signal responsive to motion of the finger touching the sensor, and a lift signal responsive to lift off of the finger from the sensor. The processor is configured to generate a scrolling signal responsive to the touch signal and the lift signal.
BRIEF DESCRIPTION OF THE DRAWING FIGURESFIG. 1A is a side plan view of a mouse typical of the prior art;
FIG. 1B is a top plan view of a mouse typical of the prior art;
FIG. 2A is a schematic view of a typical prior art rotary encoder;
FIG. 2B is a partial side plan view of a rotary disk and light detector employed by mice of the prior art;
FIG. 2C is a digital quadrature waveform generated by the rotary disk ofFIG. 2B;
FIG. 2D shows an alternative waveform to that ofFIG. 2C;
FIG. 3A is a schematic view of a rotary encoder that operates on capacitive principles rather than that which operates on optical principles as depicted inFIG. 2A;
FIG. 3B is a partial side plan view of a notched disk and related capacitance detector;
FIG. 3C is a depiction of a waveform as generated by the notched disk and capacitance detector ofFIG. 3B;
FIGS. 3D and 3E are depictions of waveforms as generated by the notched disk and capacitance detector ofFIG. 3B where the capacitance plates rotate in an opposite direction to that ofFIG. 3C;
FIG. 4 is a partial schematic side view of a capacitive rotary encoder for use herein;
FIG. 5 is a partial side plan view of a rotary encoder as an enhancement of the encoder depicted inFIG. 3A;
FIG. 6 is a partial schematic side view of a mechanism for capacitively sensing mouse motion;
FIG. 7 is a partial schematic side view of a capacitance detector and capacitance measurement circuit for use herein;
FIGS. 8A and 8B are side views of typical capacitive switches housed within a mouse enclosure;
FIG. 9 is a partial schematic side view of a scrolling wheel, capacitive rotary encoder and processor for use herein;
FIG. 10 is a partial schematic view of a further version of a capacitive scrolling control for use in the present invention;
FIGS. 11A through 11D are side and top plan views, respectively, of a mouse enclosure showing plates for capacitive sensing;
FIGS. 12A through 12E are side views of sensors mounted for use herein;
FIGS. 13A through 13D are schematic views of alternative patterns for sensors for use herein;
FIG. 14 is a top plan view of a mouse enclosure and scrolling area for use in creating the present capacitive mouse;
FIG. 15 is a graphical depiction showing total summed capacitance signal over time in employing the capacitive mouse of the present invention;
FIGS. 16A through 16C are graphical depictions of the coasting feature of the present invention;
FIG. 17 is a side view of a mouse enclosure housing the capacitive features of the present invention; and
FIG. 18 is a schematic view of a scrolling module for use as a component of the present capacitive mouse.
DETAILED DESCRIPTIONThe following description of preferred embodiments of the disclosure is not intended to limit the scope of the invention to these preferred embodiments, but rather to enable any person skilled in the art to make and use the invention.
For reference,FIG. 1A shows the elements of a conventionalprior art mouse100 in side view.Enclosure102, typically of hard plastic, forms the body of the mouse.Ball104 protrudes from the bottom ofenclosure102 through a small hole. Motion of the mouse over a flat surface causesball104 to rotate; this rotation is measured byrotary encoders106. Typically two rotary encoders are used to measure motion of the mouse in two orthogonal axes.Buttons108 form part of the top surface ofenclosure102. Finger pressure onbuttons108 is detected byswitches110 mounted below the buttons. Scrollwheel112 is mounted betweenbuttons108; its rotation is measured byrotary encoder114. Inputs fromrotary encoders106 and114 andswitches110 are combined byprocessor116 and transmitted to a host computer viacable118.
FIG. 1B shows thesame mouse100 in top view, featuringenclosure102,ball104,buttons108,scroll wheel112, andcable118.
FIG. 2A shows a typical priorart rotary encoder200. Rotation ofball202 causesshaft204 to spin, thus rotating notcheddisc206.Light emitter208 passeslight beam214 through the notches ofdisc206 tolight detector210. Asdisc214 spins, the pattern of signals fromdetector210 allowsprocessor212 to deduce the direction and speed of rotation. Note thatshaft204 is excited only by rotation ofball202 about an axis parallel toshaft204. By mounting a second rotary decoder (not shown) perpendicular torotary decoder200, rotation ofball202 about two axes, and hence motion of the mouse in a two-dimensional plane, can be detected.
FIG. 2B shows a detail view of notcheddisc206 andlight detector210.Detector210 actually contains two lightsensitive elements220 and222 spaced closely together relative to the spacing of notches224. Asdisc206 rotates in the direction indicated byarrow226, lightsensitive elements220 and222 are first both exposed to light through notch224, thenelement220 is eclipsed by the body ofdisc206, thenelement222 is also eclipsed, thenelement220 is exposed to light through adjacent notch228, thenelement222 is also exposed to light through notch228.Sensors220 and222 thus generate the digital quadrature waveform shown inFIG. 2C over time. Ifdisc206 rotates in the direction oppositearrow226, the sensors are eclipsed in the opposite order and they generate the digital waveform shown inFIG. 2D. By digitally reading the outputs oflight sensors220 and222 and decoding the quadrature signals therein, the processor can determine the direction and amount of motion ofdisc206.
In an alternate embodiment, lightsensitive elements220 and222 can be separated and placed at analogous positions within two distinct notch positions ofdisc206. This embodiment is preferable if thelight sensors220 and222 are too large to be placed closely together; the disadvantage is that it is more difficult to alignsensors220 and222 precisely relative to one another.
FIG. 3A shows arotary encoder300 that operates on capacitive instead of optical principles.Ball302 spinsshaft304 and notcheddisc306.Shaft304 anddisc306 are made of a conductive material such as metal, and the assembly consisting ofshaft304 anddisc306 is electrically grounded by groundingelement308.Capacitance detector310 measures the capacitive effects of groundeddisc306. Various methods for grounding a spinning object, such as metal brushings, are known in the art. Alternatively, onlydisc306 can be made conductive, withground308 applied directly todisc306. In yet another alternative embodiment,disc306 is capacitively coupled to a nearby grounded object. In yet another embodiment, a transcapacitance measurement may be done between the body ofdisc306 anddetector310, possibly by driving a time-varying signal intodisc306 and measuring the amplitude of coupling of that signal ontodetector310. In any case,capacitance detector310 measures the position ofdisc306 by its capacitive effects, and the resulting signals are read byprocessor312.
FIG. 3B shows a detail view of notcheddisc306 andcapacitance detector310. As in the case of the optical detector ofFIG. 2B,capacitance detector310 is formed of twoconductive plates320 and322 placed near but not touching the plane ofdisc306. When notch324 ofdisc306 is situated adjacent toplates320 and322, those plates each have a low capacitance to ground. As the body ofdisc306 moves to be adjacent to plate320 and then to plate322, the capacitance to ground of these plates rises to a higher level. Because capacitance is linearly related to the area of overlap of conductive plates, this rise of capacitance ofplate320 is linear. Asdisc306 completely coversplate320 and begins to coverplate322, the capacitance ofplate320 stays relatively constant while the capacitance ofplate322 linearly rises. Asdisc306 continues to rotate in the direction ofarrow326, the capacitance ofplate320 and then plate322 falls linearly, as depicted in the waveforms ofFIG. 3C. Ifdisc306 rotates in a direction oppositearrow326, the capacitances ofplates320 and322 instead generate the waveform ofFIG. 3D.
Those experienced in the art will recognize thatplates320 and322 may be actual metal plates, or they may equivalently be conductive regions formed in a variety of ways, including but not limited to conductive ink painted or screened on a surface or substrate, conductive material such as metal or indium tin oxide plated or otherwise disposed on a surface or substrate, or any other conductive object with at least one substantially flat portion placed in close proximity todisc306. Similarly, the conductive notcheddisc306 may be an actual notched metal disc, or it may be a notched conductive pattern formed on a disc-shaped substrate. The dielectric component of the capacitance betweenplates320 and322 anddisc306 may be an empty gap, a coating, surface, substrate, or other intermediary object, or some combination thereof whose thickness and dielectric constant yield a conveniently measurable capacitance.
Those experienced in the art will further recognize that rotary capacitive sensors are not limited to the disc configuration. Any arrangement in which an irregular conductive object rotates near a conductive sensor will work equally well. In one alternate embodiment,disc306 is extruded to form a rotating drum with a notched or patterned conductive surface, andplates320 and322 are oriented along the long dimension of the drum. The drum embodiment is bulky and mechanically more complex, but allows a larger area of capacitive overlap and hence a stronger capacitance signal. In another alternate embodiment, the notched disc could be simplified to a single “notch,” resulting in a semicircular conductive cam facing quarter-circle plates320 and322.
One way to process the capacitance signals fromplates320 and322 is to compare them against fixed capacitance thresholds. Referring toFIGS. 3D and 3E, comparingcapacitance340 againstthreshold344 yieldsdigital waveform348; similarly, comparingcapacitance342 againstthreshold346 yieldsdigital waveform350. Note thatwaveforms348 and350 ofFIG. 3E are identical in nature to the digital waveforms ofFIG. 2D. Hence, if threshold comparison is used in this manner to generate digital waveforms, these digital waveforms can be processed by aprocessor312 identical toprocessor212 of the conventional optical rotary encoder ofFIG. 2B.
Capacitance detector310 can use any of a number of methods for measuring capacitance as are known in the art. U.S. Pat. No. 5,880,411 discloses one such capacitance measuring circuit.
As in the case of the optical encoder ofFIG. 2A, note thatplates320 and322 may be placed adjacent to different notches as long as their positioning within their respective notches is maintained. However, sinceplates320 and322 do not require housings or packages outside the plates themselves, it is convenient to place them side by side mounted on a common substrate in order to ensure that they will remain aligned to each other.
One skilled in the art will observe that by examining the original analog capacitance waveforms ofFIGS. 3C and 3D, it is possible to locatedisc306 to a much finer resolution than the notch spacing. This is because at any given point in time, one of the capacitance signals is varying linearly with disc rotation while the other is constant. By tracking these linear variations,processor312 can track disc rotation at a resolution limited only by the resolution and linearity of the capacitance measurements. In the preferred embodiment, the circuits disclosed in U.S. Pat. No. 5,880,411 are used to perform these precise capacitance measurements.
Because disc rotation can be measured to much higher resolution than the notch spacing, it is possible to use much larger notches ondisc306, and correspondinglylarger plates320 and322, than are feasible for the analogous notches and sensors of the optical encoder ofFIG. 2A. Larger notches and plates allow mechanical tolerances of the assembly to be relaxed, yielding potentially lower costs. Even with larger notches and plates, a capacitive rotary encoder can produce higher-resolution data than an optical rotary encoder if a sufficiently high-resolution capacitance detector is used.Larger plates320 and322 also result in a larger capacitance signal which is easier fordetector310 to measure.
Theplates320 and322 andgrounding mechanism308, being simple formed metal pieces or plated conductive patterns, may also be less costly than the semiconductor light emitters and sensors ofFIG. 2A.
Another advantage of the capacitive rotary encoder is that it is not affected by optically opaque foreign matter, such as dirt, which may be picked up and introduced into the assembly byball306. The looser mechanical tolerances allowed by the capacitive rotary encoder may also make it more resistant to jamming by foreign matter.
FIG. 4 shows a side view of the capacitive rotary encoder, withdisc400 andplates402 and404 separate by agap406.Gap406 is drawn large for illustrative purposes, but in thepreferred embodiment gap406 is kept as small as possible to maximize the capacitance betweendisc400 andplates402 and404. Ifgap406 is small, and the tolerances of the encoder assembly are loose as previously disclosed, then movement ofdisc400 along the axis ofshaft408 will have a proportionately large effect on the width ofgap406. This variation can impact the accuracy of the capacitance measurements ofplates402 and404.FIG. 5 shows an enhancement to the arrangement ofFIG. 3A that solves this problem.
InFIG. 5,disc500 is adjacent to threeplates502,504, and506.Plates502 and504 are identical toplates320 and322 ofFIG. 3A.Plate506 is the size ofplates502 and504 combined, and is located nearplates502 and504; inFIG. 5,plate506 occupies the next notch space afterplates502 and504. In an alternative embodiment, matching could be improved by splittingplate506 into two half-plates each exactly the size ofplates502 and504. In the system ofFIG. 5, the processor computes the sum of the capacitance measurements fromplates502,504, and506. Note that the total overlap area betweendisc500 andplates502,504, and506 is constant regardless of the rotary position ofdisc500. Hence, the summed capacitance ofplates502,504, and506 should be constant. Variation in this sum indicates thatdisc500 has shifted relative toplates502,504, and506, for example, by moving along the axis as shown inFIG. 4. The processor divides each plate capacitance measurement by the summed capacitance in order to normalize the capacitance measurements. These normalized measurements are invariant of the width ofgap406 ofFIG. 4, and are suitable for use in the position computations previously discussed.
FIG. 6 shows an alternative mechanism for capacitively sensing mouse motion. This mechanism employs a rollingball602 protruding from a hole inenclosure600 similar to that of a conventional mouse. The surface ofball602 is patterned withregions604 of higher and lower conductivity. This patterning can be accomplished by forming the ball of material such as rubber of varying conductivity, or by treating the surface of the ball with conductive substances such as paint or metal. The conductive surface of the ball may be protected if necessary by a dielectricouter layer606.Capacitance detectors608 are placed in several locations proximate toball602. As the ball rolls, theconductive regions604 will move from one capacitance detector to another;processor610 correlates these signals to measure the movement ofball602. Because the capacitance measurements vary linearly asconductive region604 moves from onedetector608 to another,processor610 can interpolate in order to measure movement of the ball to very high resolution.
The system ofFIG. 6 requiresseveral sensors608 in order to ensure that at least oneconductive region604 is detectable at all times.Conductive regions604 should be as large as possible in order to maximize the capacitive signal, subject to the constraint thatdifferent regions604 should be separated by enough distance to allowindividual regions604 and the spaces between them to be resolved bydetectors608. Hence, the spaces betweenregions604 should be at least comparable to the size ofdetectors608, and theconductive regions604 should be at least a significant fraction of the size ofdetectors608.
FIG. 6 depicts a linear row ofsensors608 curved around the surface ofball602. Such an arrangement can detect rolling of the ball in one dimension; the example ofFIG. 6 would detect the rolling resulting from motion of the mouse alongaxis612. In the preferred embodiment, other sensors (not shown) are arranged in a row perpendicular to the row ofsensors608 in order to measure motion of the mouse in two dimensions.
In one embodiment, the conductive regions in the ball are grounded to facilitate capacitance measurements by simple conductive plates. However, grounding the conductive regions of the ball may be impractical, so in the preferred embodiment,capacitance detectors608 measure transcapacitance.
FIG. 7 shows one simple way to measure transcapacitance. Thecapacitance detector700 consists of twoplates702 and704.Plate702 is connected to ground, andplate704 is connected to acapacitance measurement circuit706. Proximity to an electrically floatingconductor708 withinball710 creates acapacitive coupling712 fromplate702 toconductor708, and acapacitive coupling714 fromconductor708 toplate704, hence effectively couplingplate702 to plate704 through two series capacitances. Those experienced in the art will recognize that many other configurations ofplates702 and704 are possible, such as interdigitated lines or concentric circles and toroidal shapes. In still another embodiment ofcapacitance detector700,plate702 could be driven with a time-varying signal which is capacitively coupled ontoplate704 and detected bycircuit706.
The motion sensor ofFIG. 6 requires even fewer moving parts than that ofFIG. 3, and thus can lead to an even cheaper and more physically robust mouse. However, the system ofFIG. 6 has the disadvantage of requiring more complex processing inprocessor610.
Other methods for detecting mouse motion are known in the art, such as the optical methods of U.S. Pat. No. 4,546,347 (Kirsch) and U.S. Pat. No. 5,907,152 (Dandiker et al.). Fully solid-state optical motion detectors would pair well with the capacitive button and scrolling controls of the present invention to form an entirely solid-state optical/capacitive mouse.
Mice conventionally include one or more buttons as well as a motion detector. Referring back toFIG. 1,button108 is typically linked to amechanical switch110. By pressing down on the surface ofswitch108, the user closesswitch110. Mechanical switches have various well known disadvantages. Since they have moving parts, mechanical switches can fail over time or with rough handling. Also, mechanical switches require a certain threshold of pressure for activation, which can tire the user with repeated use.
Mechanical switches can be replaced by capacitive sensors in several ways.FIG. 8A shows one type of capacitive switch that is well-known in the art.Enclosure800, for example a mouse enclosure, may be shaped similarly to that of a conventional mouse, but with no moving parts in its top surface.Conductive plate802 is placed on or near the surface of the enclosure, preferably covered by aprotective dielectric layer806. In an embodiment,capacitance measurement circuit804 monitors the capacitance ofplate802. When a finger (not shown) touchessurface region806, the capacitance to ground ofplate802 increases beyond a threshold set bymeasurement circuit804. When no finger is present, the capacitance to ground ofplate802 is below the threshold. By comparing the capacitance ofplate802 to the threshold,circuit804 can generate a digital signal which is equivalent to the signal produced by a mechanical switch.
The system ofFIG. 8A implements a mouse button which requires zero activation force; indeed, depending on the threshold setting, it could even be sensitive to mere proximity of the finger. Although this mouse button solves the problem of tiring the finger during repeated activations, it introduces the converse problem of tiring the finger during periods of inactivity, since the finger must not be rested againstsurface806 without accidentally activating the button.
FIG. 8B shows a second type of capacitive switch, also well-known in the art.Enclosure820 includes a separatemovable button portion822 as in a conventional mouse. Instead of a mechanical switch beneathbutton822, there is aconductive plate826 and some sort ofspring mechanism824. A variety ofmechanisms824 are usable and well-known, including but not limited to metal springs, compressible foam, or single-piece enclosures with buttons made of springy material.Spring mechanism824 may optionally also include a tactile feedback means to impart the familiar clicking feel to button activations. A secondconductive plate828 is mounted beneathplate826 so that pressure onbutton822 bringsplate826 measurably closer to plate828, thus increasing the capacitance betweenplates826 and828. Capacitance measuringcircuit830 detects this change in capacitance to form a button signal.
Because the system ofFIG. 8B works by measuring the capacitance betweenplates826 and828, these plates do not need to make electrical contact in order to activate the button. Indeed, these plates must be kept out of electrical contact in order forcapacitance measuring circuit830 to operate properly. Many straightforward ways are known to separateplates826 and828, including but not limited to an insulating surface onplate826,plate828, or both plates, or an insulating compressible foam placed between the plates.
The system ofFIG. 8B is very similar to a conventional mechanical switch, but it is more resistant to dirt and wear because button activation does not require an electrical contact to be made.
Capacitance measuring circuits804 and830 may use any of a variety of well-known capacitance measuring techniques. In the preferred embodiment, a circuit like that disclosed in U.S. Pat. No. 5,880,411 is used.
Many mice also include a scrolling mechanism. This mechanism typically employs a rotating wheel, an isometric joystick, or a set of directionally arranged buttons; thescrolling mechanism112 is typically mounted between twomouse buttons108 as shown inFIG. 1B.
FIG. 9 shows one way to measure a scrolling command capacitively. Ascrolling wheel902 is mounted inmouse enclosure900, seen in side view. The wheel appears to the user to be the same as the wheel of the conventional mouse ofFIG. 1A and 1B. Rotation of the wheel is measured by capacitiverotary encoder904 andprocessor906 similar to those ofFIG. 3A and 3B. The capacitiverotary encoder904 can be mounted directly on the axis of scrollingwheel902 as shown inFIG. 9, orwheel902 can be mechanically linked to a separate rotary encoder mechanism elsewhere inenclosure900.
FIG. 10 shows another capacitive scrolling control. A scrollingknob1002 protrudes frommouse enclosure1000.Knob1002 is connected bystick1004 toconductive plate1006 and tospring mechanism1008. Depending on the stiffness ofspring1008,knob1002 may act as either a rocking control or an isometric joystick.Conductive plates1010 and1012 are mounted nearplate1006, andcapacitance measuring circuit1014 measures the capacitances betweenplate1010 andplate1006, and betweenplate1012 andplate1006. Whenknob1002 is pressed in a forward or backward direction,plate1006 is deflected slightly to produce a measurable change in the capacitances ofplates1010 and1012. By comparing the capacitances ofplates1010 and1012,circuit1014 can detect this forward or backward deflection to produce a scrolling command. Also, by noting an increase in capacitance of bothplates1010 and1012 at once,circuit1014 can detect downward pressure exerted onknob1002. Many conventional mice use a downward deflection of the scrolling control as an additional command signal, such as the activation of a third mouse button.
By placing two additional plates along an axis perpendicular to the axis ofplates1010 and1012, it is possible to measure deflection ofknob1012 in three dimensions. Sideways deflection ofknob1012 can be interpreted as a command for horizontal scrolling, or panning Forward and backward deflection can be interpreted as vertical scrolling, and downward deflection can be interpreted as an additional mouse button or other special command.
In an alternate embodiment,plates1010 and1012 are situated aboveplate1006 so that pressure onknob1002 causesplate1006 to deflect away fromplates1010 and1012, and the measured capacitance onplates1010 and1012 to decrease with pressure instead of increasing. Those skilled in the art will recognize that the processing necessary for this embodiment is identical to that required for the embodiment ofFIG. 10 except for a change of sign.
The systems ofFIGS. 9 and 10 share the disadvantage that they are still mechanical devices with moving parts. For greatest robustness and sensitivity and lowest cost, a truly solid state solution to scrolling is preferable.
FIG. 11A shows a scrolling control that operates directly on capacitive sensing principles.Mouse enclosure1100 contains an array ofconductive plates1102 connected to aprocessor1104 that includes capacitance measuring circuits.Plates1102 are insulated from the user's finger bysurface1106. In the preferred embodiment, the array ofplates1102 is placed in between twomouse buttons1108 as shown inFIG. 11B. Many alternate embodiments in which the scrolling control is placed elsewhere are possible, such as the embodiment ofFIG. 11 C in which the scrolling control is mounted on the side ofmouse enclosure1100 for access by the user's thumb. Themouse buttons1108 ofFIGS. 11B and 11C could be capacitive buttons as previously disclosed, or conventional mechanical switches or any other suitable type of button.
FIG. 11D shows yet another configuration, in whichscrolling sensors1102 are placed on top of aconventional mouse button1108; pressing down onbutton1108 without substantially moving the finger produces a button click, while lightly touchingbutton1108 and then moving the finger generates scrolling.
Preferably,plates1102 are numerous and spaced closely together so as to allow interpolation of the finger position to greater resolution than the plate spacing. In one preferred embodiment, nine plates are used spanning a distance of approximately one inch. U.S. Pat. No. 5,880, 411 discloses a preferred method for measuring the capacitances of an array of sensors and interpolating the finger position from those measured capacitances. Many other methods are possible and well-known in the art, such as that of U.S. Pat. No. 5,305,017 to Gerpheide.
Once the finger position amongplates1102 is known, motion of the finger along the axis ofplates1002 can be measured by comparing finger positions at successive times.Processor1104 generates a scrolling signal of a certain direction and distance when a finger motion of a corresponding direction and distance is measured. The effect as observed by the user is as if the user were rolling a wheel likewheel902 ofFIG. 9 by moving the finger forward and backward on the top edge of the wheel. Instead, the user moves the finger forward and backward alongsensor surface1106 to produce the identical scrolling signals.
In any scrolling mouse, but particularly in a capacitive scrolling mouse, it may be desirable to provide for different regimes of low-speed and high-speed scrolling in order to account for the fact that thescroll surface1106 is much shorter than a typical scroll bar in a typical graphical user interface. A simple way to provide for different speed regimes is to use the technique commonly known as “acceleration” or “ballistics” when applied to mouse motion signals. In this technique, very small finger motions translate to disproportionately small scroll signals, and very large finger motions translate to disproportionately large scroll signals.
In the preferred embodiment,processor1104 measures the total amount of finger signal as well as the finger position, and generates a scrolling signal only when sufficient finger signal is present. Otherwise, the scrolling signal when no finger was present would be ill-defined, and the mouse would be prone to undesirable accidental scrolling. In the preferred embodiment,processor1104 compares the total summed capacitance on allsensors1102 against a threshold to determine finger presence or absence; in an alternate embodiment,processor1104 instead compares the largest capacitance signal among allsensors1102 against a threshold. The threshold should be set high enough so that only deliberate finger actions result in scrolling. If the threshold is set too low, the mouse may scroll in response to mere proximity of the finger, in general an undesirable feature.
There are many ways to mountsensors1102 undersurface1106 or to otherwise integrate the sensors intoenclosure1100. Some of these ways are depicted inFIGS. 12A through 12E. Those experienced in the art will realize that many other mounting schemes are possible, and that the particular choice of mounting scheme does not alter the essence or the basic operation of the invention.
InFIG. 12A, scrollingsurface1202 is an uninterrupted region of thetop layer1201 ofenclosure1200. In the embodiment illustrated inFIG. 12,top layer1201 includes an external surface (front surface)1207 and an internal surface (back or underside surface)1203.Sensors1204 are affixed to theinternal surface1203 ofenclosure1200, for example using adhesive or otherintermediary substance1206.Adhesive1206 could be eliminated by the use of a self-adhesive sensor material1204 such as conductive paint. Wires orother conductors1205connect sensors1204 toprocessor1208.
InFIG. 12B,sensors1204 are disposed on asubstrate material1206 which is then affixed to theinternal surface1203 ofenclosure1200.Sensors1204 might be composed of conductive ink, indium tin oxide, metal foil, or any other conductive material.Substrate1206 might be polyester film, plastic, glass, or any other flexible or bendable material on which conductive sensors can be disposed. In the example ofFIG. 12B,substrate1206 is shown extending downwardly (e.g., through bending) and away fromenclosure1200 to carry the conductive signals fromsensors1204 toprocessor1208.
InFIG. 12C, the material which forms thetop layer1201 ofenclosure1200 in or near scrollingregion1202 has been made thinner than normal in order to reduce the distance between, and thereby increase the capacitive coupling between,sensors1204 and the finger. Additionally,sensors1204 have been disposed on the opposite side of substrate1206 (as compared to the embodiments shown inFIGS. 12A and 12B) in order to increase the proximity of the sensors to the finger. To strengthen thetop layer1201 of the enclosure,solid backing plate1210 can optionally be placed behind thesensors1204.Layer1210 may also be made conductive and electrically grounded in order to isolatesensors1204 from interference from other circuits within the mouse. A similar grounded shield may be used in any of the other sensor arrangements disclosed herein.
InFIG. 12D,substrate1206 extends from the inside ofenclosure1200, with an intermediate portion ofsubstrate1206 extending out throughhole1212 to the external surface (outside surface) ofenclosure1200. In this example,substrate1206 itself forms the protective dielectric layer associated with the scrollingsurface1202 betweensensors1204 and the finger.Hole1212 may be protected and disguised in various ways, such as by combininghole1212 with the opening around the edge of a mechanical mouse button.
InFIG. 12E,sensors1204 are embedded directly into the material ofenclosure1200, for example in the form of wires or foil strips encased in plastic.
Whensensors1204 are disposed on asubstrate1206, it is convenient to use an extension ofsubstrate1206 to carry the sensor signals toprocessor1208, as shown inFIGS. 12B,12C, and12D. In these cases,sensors1204 and their associated wiring may be patterned onsubstrate1206 using conductive ink or other suitable material.FIGS. 13A to 13D show several of the many possible patterns.
InFIG. 13A,substrate1300 extends beyond the area ofsensors1302 on one side (e.g., to the right). Thisside extension1306 forms a carrier for the sensor signals carried by the wires orconductors1304 to aprocessor1308.Processor1308 may be mounted to the side ofsensor area1302 as shown, or it may be mounted beneathsensor1302 or in another location, with abendable extension1306 bending, folding or warping as it leads away fromsensor1302.
InFIG. 13B,wires1304 are shown as being bent at90 degrees andextension1306 leads away along the length of the area ofsensors1302.
FIG. 13C is similar toFIG. 13B, butsensors1304 leave the area ofsensors1302 on both sides in order to balance the extension ofsubstrate1300 to the sides of the area ofsensors1302.
InFIG. 13D, two layers of conductive material are used with an insulating layer or substrate therebetween. The first conductive layer containssensors1302. The second conductive layer containsconductors1304 which carry the sensor signals and which extend in a direction perpendicular tosensors1302.Vias1310 penetrate the insulating layer or substrate to connectsensors1302 to signalwires1304. Incrossings1312 ofwires1304 oversensors1302 without vias, the two conductive layers are electrically isolated although there may be some capacitive coupling thatprocessor1308 may take into account. The sensor design ofFIG. 13D may be more expensive due to its use of additional layers, but it avoids excess extension ofsubstrate1300 around the area ofsensors1302. Such extension may be undesirable for design or aesthetic reasons, in addition to providing opportunities for undesirable capacitive coupling between the finger andwires1304 when the finger touches near but not directly in the area ofsensors1302. The latter undesirable capacitive coupling can also be remedied by the addition of a grounded shield over the exposedwires1304, as shown byregion1314 ofFIG. 13B.
Yet another embodiment of the capacitive scrolling control is shown inFIG. 14.Mouse enclosure1400 includes a two-dimensional scrolling area1402 preferably disposed betweenmouse buttons1408. Scrollingarea1402 includes first plurality ofsensors1404 disposed in one direction, and a second overlapping plurality ofsensors1406 disposed in a substantially perpendicular direction to form a two-dimensional matrix. Each plurality of sensors is processed using methods analogous toFIGS. 11 through 13; the position results from the two pluralities are combined to form the complete finger location in two dimensions.
Two-dimensional capacitive touch sensors, or touch pads, are well known in the art. In the preferred embodiment, the methods of U.S. Pat. No. 5,880, 411 are used.FIG. 2 of the '411 patent illustrates a diamond pattern forsensor matrix1402 which is preferred due to various advantages disclosed in that patent. Many other sensing techniques and sensor geometries are known in the art.
Once the finger position in two dimensions is known, finger motion in the horizontal and vertical directions can be measured by comparing finger positions at successive times. Horizontal finger motion translates to horizontal scrolling, or panning Vertical finger motion translates to vertical scrolling. In one embodiment, diagonal finger motion translates to simultaneous horizontal and vertical scrolling. In an alternate embodiment, the horizontal and vertical motion signals are compared to discover whether the finger motion is primarily horizontal or primarily vertical, and the corresponding type of scrolling is applied.
Scrolling wheel mice like that of U.S. Pat. No. 5,530,455 typically contain an additional switch to sense when the wheel is pressed down by the user. This switch generates a signal similar to a third mouse button signal for enabling additional scrolling or other features in host software. A comparable switch could be mounted beneath the capacitive touch sensors ofFIGS. 11 through 14, but other methods are preferred in order to avoid the cost and reliability problems inherent in switches.
One way to simulate a third mouse button in a capacitive scrolling control is to decode tapping gestures using the various methods disclosed in U.S. Pat. No. 5,880,411. In the most simple case, basic finger taps are decoded and translated into simulated clicks of the third mouse button.FIG. 15 shows the total summed capacitance signal over time, and the corresponding third button signal resulting from tap detection. The '411 patent discloses many additional refinements for tap detection on capacitive touch sensors, many of which are suitable for application to scrolling controls.
A second way to simulate a third mouse button is to introduce an additional touch sensor plate which forms a capacitive button as disclosed inFIGS. 8A or8B.
Arrayed capacitive touch sensors, particularly two-dimensional sensors like that ofFIG. 14, can resolve numerous additional types of input that more specialized sensors like wheels and isometric joysticks cannot. One example is the use of multiple fingers to activate special modes or user interface commands; U.S. Pat. No. 5,880,441 discloses one embodiment of multi-finger sensing. Another example is graphic gestures, where looping motions and other finger motions that are not entirely horizontal or vertical can be interpreted as special user interface commands. Yet another example is special designated zones in which finger motion or tapping invokes special behaviors.
Because the capacitive scrolling control feels similar to a scrolling wheel to the user, other techniques may be employed to strengthen the wheel analogy. One such technique is “momentum” or “coasting,” in which scrolling behavior is adjusted based on the velocity of finger motion as the finger lifts away from the scroll sensor; that is, the scrolling speed may be determined based on the instantaneous velocity of the finger at the point the finger lifts from the scroll surface.
FIGS. 16A and 16B illustrate the basic coasting feature. Each figure shows the finger presence or absence, the computed finger motion, and the resulting scrolling signal generated by the mouse. For simplicity, motion and scrolling in only one dimension are considered as in the case ofFIG. 11; the two-dimensional scrolling ofFIG. 14 leads to a straightforward generalization ofFIG. 16. Note that the finger motion is undefined when the finger is absent; inFIGS. 16A and 16B, the motion is plotted as zero when the finger is absent for purposes of illustration.
InFIG. 16A, the finger touches the scrolling sensor, moves back and forth to generate a corresponding back-and-forth scrolling signal, then comes to a complete stop before lifting. When the processor observes a zero or near-zero velocity as the finger lifts, i.e., when the finger is stationary immediately prior to lifting, the processor ceases all scrolling activity; coasting does not occur.
InFIG. 16B, the finger executes the same scrolling motions, but then moves again and lifts while still moving. When the processor observes that the velocity was substantially non-zero as the finger began lifting, the processor continues scrolling in a direction and speed determined by the final velocity of the finger upon lifting. The effect as seen by the user is that the imaginary scroll wheel is left spinning, or coasting—much like a roulette wheel continues to rotate after manually spinning it. In the preferred embodiment, the coasting speed and direction are equal to the scrolling speed and direction just before the finger lifted, though in alternate embodiments, the coasting speed could be constant or the coasting speed and direction could be some other function of the final scrolling speed and direction.
To terminate coasting, the user simply returns the finger to the scrolling control as seen inFIG. 16B. No special processing is needed to accomplish this aspect of coasting: As soon as the finger returns to the scrolling control, the coasting signal is replaced by fresh motion signals, which are zero until the finger actually moves on the control. The effect as seen by the user is that the imaginary spinning scroll wheel is halted as soon as the finger is pressed on it. Coasting is a valuable aid to long-distance scrolling through large documents.
FIG. 16C shows an additional embodiment or implementation of the coasting feature, wherein friction is simulated thereby having the coasting speed slowly decay to zero. Much like a spinning roulette wheel loses its momentum and slows down to a stop.FIG. 16C shows an alternate scrolling signal to that ofFIG. 16B in which friction slows the coasting effect over time.
The user can still halt the coasting before it has come to a natural stop by touching the finger back to the scrolling control.
Some mice offer other features in addition to motion, two buttons, and scrolling. Many of these features are also well suited to a capacitive implementation. One example is additional buttons for special functions such as Internet browsing. Another example is additional scroll-like functions such as a separate “zoom” control. Still another example is a general hand proximity sensor on the mouse enclosure that allows the mouse and associated software to tell whether or not the user's hand is gripping the mouse. Those experienced in the art will recognize that the various types of capacitive sensors, buttons, rotary, linear and two-dimensional, are appropriate for a wide variety of applications beyond those specific examples disclosed here.
Referring back toFIG. 1, any combination of one or more of themotion sensors106,button sensors110, scrollingsensors114, and any additional sensors can be implemented by capacitive methods as disclosed herein. In typical mice, the signals from all these types of sensors, whether capacitive, mechanical, optical or otherwise are combined inprocessor116 to produce a mouse signal to be sent to the host computer. Standard protocols are well known in the art for sending motion, button, and scrolling signals from a mouse to a host computer. These same protocols may be used when one, several, or all of the sensors are implemented by capacitive techniques. Thus, the capacitive mouse of the present invention is fully interchangeable with conventional mice with no change to host mouse drivers or other system-level facilities.
It is possible and may be desirable to construct a mouse that uses a combination of capacitive, mechanical and other sensing techniques. For example, a capacitive scrolling sensor could be added to an otherwise conventional mechanical mouse. Or, a capacitive motion sensor could be used on a mouse with mechanical buttons and no scrolling control at all.
If several or all sensor functions of the mouse are implemented capacitively, it may be possible to use a single capacitive sensing chip for all capacitive sensing functions. Thus, for example, if capacitive sensing is used on the mouse for scrolling, then it may cost little more to implement the motion sensor capacitively as well using additional input channels of the same capacitance measuring chip.
It is possible to purchase mouse processor chips that perform all of the tasks ofprocessor116 or a conventional mouse. These chips generally accept motion and scrolling inputs in quadrature form as shown inFIGS. 2C and 2D, and the buttons are implemented as switches which alternately drive an input pin to a high or low voltage.
FIG. 17 shows how acapacitive mouse1700 can be built using a conventionalmouse processor chip1702 in conjunction with acapacitance measuring chip1704.Ball1706 drives capacitivemotion sensor1708, whose sensing plates connect tochip1704. Scrollingsensors1710 also connect to chip1704, as do the button sensors (not shown).Chip1704 computes motion and scrolling signals using the techniques disclosed herein, and then generates quadrature signals as outputs with timing and characteristics matching those produced by a true rotary sensor such as that ofFIG. 2A.Chip1702 then converts these artificial quadrature signals into standard mouse protocols. If quadrature is not appropriate,chips1704 and1702 could equally well use any other intermediate form for transmitting motion data.Chip1704 also measures the signals from the capacitive mouse buttons, and drives its digital output pins high or low based on the observed button capacitances.Chip1702 reads these digital button signals as if they came from mechanical switches. The arrangement ofFIG. 17 is not as cost-effective as a design with a single chip that does all the tasks, but it may greatly simplify the design of a new mouse using a new protocol or other features not yet supported by standard capacitive sensing chips.
Yet another alternative is to perform only rudimentary sensor processing on the mouse, producing an intermediate form such as the quadrature output bychip1704 ofFIG. 17. These signals can then be sent to a host computer for final processing, thus relieving some of the load from the low-cost mouse hardware. Another variation of this scheme is to send finger position data instead of fully processed scrolling motion data for a capacitive scroll sensor.
FIG. 18 shows a scrolling module designed to be used as a component in a mouse design.Circuit board1800 includes an array ofsensors1802 as well as acapacitive sensing chip1804.Connector1806 sends out quadrature signals compatible with conventional rotary encoders. Similarly, a self-contained rotary encoder module could be constructed using capacitive sensors. Using these modules, an industrial designer could construct the mouse ofFIG. 17 using only standard components, without requiring any expertise in capacitive sensing.
As any person skilled in the art will recognize from the previous description and from the figures and claims, modifications and changes can be made to the preferred embodiments of the invention without departing from the scope of the invention defined in the following claims.