WO2010134349A1 - 触感処理装置 - Google Patents
触感処理装置 Download PDFInfo
- Publication number
- WO2010134349A1 WO2010134349A1 PCT/JP2010/003429 JP2010003429W WO2010134349A1 WO 2010134349 A1 WO2010134349 A1 WO 2010134349A1 JP 2010003429 W JP2010003429 W JP 2010003429W WO 2010134349 A1 WO2010134349 A1 WO 2010134349A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- tactile
- tactile sensation
- movement
- unit
- physical
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/016—Input arrangements with force or tactile feedback as computer generated output to the user
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
- G06F3/014—Hand-worn input/output arrangements, e.g. data gloves
Definitions
- the present invention relates to an apparatus, a method, and a program for numerically measuring and reproducing or searching for a tactile sensation when a person touches an object.
- the tactile sensation is a sensation when a person touches the object, for example, a sensation related to the hardness or surface characteristics of the object such as “hard / soft” or “smooth / rough”.
- the audio-visual system is a non-contact type sensory system that humans feel by receiving light and sound waves, while the tactile system is a contact-type sensory system in which a sensation occurs only when an object is touched.
- Quantification of tactile sensation is performed, for example, by sensory evaluation for collecting human tactile responses and factor analysis that numerically expresses tactile sensation by analyzing collected data, as disclosed in Non-Patent Document 1.
- a question sheet 1501 as shown in FIG. 2 is presented to the subject, and the tactile sensation when touching the object is answered numerically for each question item.
- the question item is composed of adjective pairs having opposite meanings. For example, the “hard-soft” degree is answered in seven levels.
- Non-Patent Document 1 subjects were made to respond to tactile sensations of 20 types of objects using 12 types of question items.
- Factor analysis is one of the methods of multivariate analysis, and is a method of analyzing individual components on the assumption that the observation data is a composite quantity.
- sensory evaluation results can be summarized into four factors, which are factors relating to a feeling of roughness, a feeling of coolness / warmth, a feeling of dryness / humidity, and a feeling of hardness / softness, respectively.
- the subject's tactile sensation can be quantitatively described as a tactile sensation feature vector 1602 in a four-dimensional feature amount space 1601 having the four factors as axes.
- the tactile feature vector 1602 is a quantitative human response characteristic obtained through a sensory evaluation experiment, and corresponds to a psychological quantity.
- Non-Patent Document 2 tactile sensing is performed by associating a physical characteristic value of an object with the tactile feature vector.
- the purpose of Non-Patent Document 2 is to objectively evaluate the texture of the cloth.
- the texture characteristic Hk of the cloth is expressed by a linear combination of the mechanical and surface characteristics xi of the cloth.
- xi is 16 kinds of physical measurement values consisting of tensile property value, bending property value, shear property value, compression property value, surface property value, thickness property value, and weight property value
- Hk is strain, slime , Bulge, shaving and beam.
- Xi with a bar at the top is an average value of a plurality of samples
- ⁇ i is a standard deviation value of the plurality of samples
- Ck0 and Cki are constants.
- the constants Ck0 and Cki are calculated by regression analysis from the mechanical / surface characteristics xi and the texture characteristics Hk of the cloth obtained from a plurality of cloth samples.
- the object is set on a measuring instrument, and the measuring instrument deforms (tensile, bends, shears, compresses, etc.) the object as necessary to measure the target physical property value. .
- the texture characteristic Hk is obtained by answering the strength that the subject touches the object in the sensory evaluation experiment and feels for each texture.
- the process of calculating the constants Ck0 and Cki is a so-called “learning process”, and after completing this process, Equation 1 is completed. Thereafter, the process proceeds to the “execution step”, where the mechanical and surface characteristics xi of the cloth with respect to an unknown object are measured, and Equation 1 is calculated to estimate the texture characteristic Hk.
- the texture characteristic Hk is a quantitative human response characteristic obtained through a sensory evaluation experiment, which corresponds to a psychological quantity. Therefore, Formula 1 is a physical psychological conversion formula for converting a physical quantity into a psychological quantity, and converts a physical quantity as a mechanical / surface characteristic xi of a cloth into a psychological quantity as a texture characteristic Hk.
- FIG. 4 is a block diagram showing the flow of processing in the prior art for the tactile quantification and tactile sensing described above.
- the tactile sensation of the subject 1703 is quantitatively expressed using a plurality of learning objects 1702.
- the subject 1703 answers the tactile sensation of the learning object 1702 according to the sensory evaluation means 1704.
- This answer result is subjected to multivariate analysis by factor analysis means 1705, the factor is analyzed, and this factor group is output as a learning material tactile feature vector F.
- the tactile feature vector 1602 in FIG. 3 used for the description of Non-Patent Document 1 corresponds to the learning material tactile feature vector F in FIG.
- physical property values of a plurality of learning objects 1702 are measured by the object physical measurement unit 1706.
- This measurement result is output as a learning material physical property value vector Ps.
- the mechanical / surface property xi of the number 1 used in the description of Non-Patent Document 2 corresponds to the learning material physical property value vector Ps in FIG.
- the physical psychological conversion calculating unit 1707 calculates a function M for converting the learning material physical characteristic value vector Ps into the learning material tactile feature vector F as shown in Equation 2.
- the means for calculating the constants Ck0 and Cki of Equation 1 used in the description of Non-Patent Document 2 corresponds to the physical psychological conversion calculating means 1707.
- a matrix is used as the function M
- Equation 2 is a matrix conversion formula.
- the execution step 1708 the physical property value of the test object 1709 whose tactile sensation is to be measured is measured by the object physical measurement unit 1706 to obtain a reference material physical property value vector Pt.
- the physical psychology conversion unit 1710 converts the reference material physical property value vector Pt into the estimated tactile feature vector F ′ according to Equation 3 using the function M calculated in the learning step 1701.
- the tactile sensation display generates a tactile sensation by deforming the human skin with an actuator.
- an ultrasonic vibrator is used as an actuator, and the feeling of roughness and softness are controlled using the squeeze effect of the ultrasonic vibrator.
- the squeeze effect is a phenomenon in which pressure is generated in the fluid between two objects when the two objects approach rapidly, and fluid lubrication acts.
- FIG. 5 shows the relationship between the structure of the tactile sensation display unit 1801 and the feeling of roughness 1803, hardness / softness 1804, and friction feeling 1805 which are tactile sensations of the person 1802.
- the tactile display unit 1801 is in contact with the finger of the person 1802 and interacts with it.
- a portion where the tactile sensation display unit 1801 and the person 1802 interact with each other is defined as an interaction unit 1806, which is indicated by a dotted square.
- the tactile sensation display unit 1801 acts on the person 1802 with vibration 1807 to present a feeling of roughness 1803.
- the vibration 1807 is generated as the sum of the amplitude-modulated stationary component 1813 and the unsteady component 1809 of the ultrasonic vibration unit 1808, and the unevenness of the height corresponding to the unsteady component of the amplitude-modulated wave is several tens of times the vibration amplitude.
- the roughness 1803 is controlled by the non-stationary component 1809 by utilizing the perception of.
- the vibration frequency transmitted to the finger is high.
- the vibration frequency transmitted to the finger is low. Therefore, since the vibration frequency transmitted to the human finger is proportional to the speed of the human finger, the finger speed 1810 is measured by the position sensor unit 1811 and used to control the unsteady component 1809.
- the tactile sensation display unit 1801 acts on the person 1802 with a force distribution 1812 in order to present a softness and softness 1804.
- the force distribution 1812 is controlled by a steady component 1813 of amplitude modulation of the ultrasonic vibration unit 1808.
- the steady component 1813 also affects the vibration 1807. Therefore, the influence on the roughness 1803 is corrected by adjusting the amplitude ratio of the steady component 1813 and the unsteady component 1809.
- the tactile sensation display unit 1801 acts on the person 1802 with a frictional force 1814 in order to present a feeling of friction 1805. Since the squeeze effect of the ultrasonic vibrator lowers the friction coefficient, it is difficult to control the friction feeling independently using only the ultrasonic vibrator. Therefore, the change in the friction characteristic of the ultrasonic transducer is corrected by the presentation of the tangential force 1816 by the force sense presentation unit 1815.
- the tangential force 1816 is calculated according to the finger velocity 1810 and finger position 1817 detected by the position sensor unit 1811 and the normal force 1819 of the human finger detected by the force sensor unit 1818.
- a static friction force corresponding to the displacement from the initial contact position is presented.
- the ratio of the tangential force to be presented and the normal force applied to the tactile sensation presentation unit exceeds the static friction coefficient presented by the force sense presentation unit 1816, the dynamic friction force is presented to the person 1802.
- the tactile sensation display unit 1801 causes the vibration 1807, the force distribution 1812, and the frictional force 1814 to act on the person 1802 with the ultrasonic vibration unit 1808 and the force sense presentation unit 1815, and feels the roughness 1803 and the softness / softness. 1804 and a feeling of friction 1805 are presented to the person 1802. At this time, the friction coefficient is reduced due to the squeeze effect of the ultrasonic vibrator, and it is difficult to control the feeling of friction with only the ultrasonic vibrator. Therefore, the tangential force 1816 generated by the haptic presentation unit 1815 reduces the friction feeling. Apply correction.
- the tangential force 1816 generated by the force sense presentation unit 1815 is calculated based on the finger speed 1810, the finger position 1817, and the normal force 1819 obtained from the position sensor unit 1811 and the force sensor unit 1818.
- FIG. 6 is a block diagram of a tactile sensation processing apparatus 2100 configured by the inventor of the present application virtually combining the tactile quantification and tactile sensing described in FIG. 4 and the tactile display described in FIG.
- the tactile sensor unit 2101 is the same as that shown in FIG. 4, and the object physical measurement unit 1706 measures the physical property value of the test object 1709 and outputs a reference material physical property value vector Pt.
- the physical psychological conversion unit 1710 has a function P for physical psychological conversion in advance through the learning process 1701 shown in FIG. 4, and converts the reference material physical characteristic value vector Pt into the estimated tactile feature quantity vector F ′. .
- a person 1802 touches the actuator unit 2102 to feel a tactile sensation.
- the actuator unit 2102 corresponds to the ultrasonic vibration unit 1808 and the force sense presentation unit 1815.
- the actuator control unit 2103 drives the actuator unit 2102. In the case of FIG. 5, it corresponds to an electric means for vibrating the ultrasonic vibration unit 1808 and an electric means for driving the force sense presentation unit 1815.
- the psychophysical conversion unit 2104 converts the estimated tactile feature vector F ′ into an actuator control signal D ′ so that the tactile sensation acquired by the tactile sensor unit 2101 can be reproduced by the tactile sensation display unit 1801.
- the function Q is a psychophysical conversion function that converts the estimated tactile feature quantity vector F ′, which is a psychological quantity, into an actuator control signal D ′, which is a physical quantity.
- This function Q is determined by the input / output characteristics of the tactile display unit 1801. That is, it is determined by the relationship between the actuator control signal D ′ that is an input to the tactile display unit 1801 and the tactile sensation Fo sensed by the person 1802 that is the output of the tactile display unit 1801, which is expressed by the following mathematical expression.
- Equation 4 is rewritten as follows.
- an actuator control signal D ′ that allows the person 1802 to feel the tactile sensation of the test object 1709 can be calculated.
- a tactile sensation when a person touches an object is sensed and transmitted as a tactile sensation feature vector over a network, and the tactile sensation can be reproduced on a tactile display.
- the tactile sensation feature quantity vector 1602 always includes tactile sensations that are not felt by humans, and thus is excessive as information and has a problem of diluting important tactile information.
- Non-Patent Document 4 a person usually obtains various tactile sensations by changing “how to touch”, that is, “contact state between a hand and an object”.
- the softness 1804 is a tactile sensation obtained mainly by pushing the object, while the roughness 1803 and the friction sensation 1805 are mainly obtained by tracing the object. It is a tactile feeling.
- the friction feeling 1805 is a tactile sensation generated from a frictional force 1904 acting in a tangential direction with the object 1903 by the tracing operation 1902 of the finger 1901 as shown in FIG. Therefore, as shown in FIG. 7B, the frictional force 1805 is not felt by the pushing operation 1905 that is perpendicular to the tangent to the object 1903.
- the soft feeling 1804 and the friction feeling 1805 are tactile sensations that are individually sensed depending on the movement of the hand, and a person operates to push in an object in order to know the hard feeling 1804. In this case, the friction feeling 1805 cannot be felt.
- a human finger includes a plurality of tactile receptors having different frequency sensitivity distributions.
- Merkel cells, Meissner bodies, and Patini bodies have sensitivity distributions as shown in FIG.
- the vertical axis 2001 in FIG. 8 represents the firing threshold value of the tactile receptor, and is the threshold value of the push-in amplitude at which the tactile receptor fires when the finger is pushed into the object.
- the horizontal axis 2002 in FIG. 8 represents the frequency of the finger pressing operation.
- the pachinko body 2003 has the highest sensitivity among the three tactile receptors, and ignites with a pushing amplitude of 2 ⁇ m for vibrations of about 80 Hz. For example, when the vibration frequency is lowered to 10 Hz, the sensitivity is lowered and the ignition threshold is increased to 100 ⁇ m.
- the Pachiny body 2003 has a sensitivity distribution corresponding to the frequency, with 100 Hz being the peak sensitivity.
- the Merkel cell 2004 and the Meissner body 2005 have different sensitivity depending on the frequency, and have a frequency sensitivity component.
- the push operation When comparing the human pushing operation and the drag operation, generally, the push operation has a low frequency and the drag operation has a high frequency. That is, it is natural to interpret that the frequency at which the skin vibrates by tracing the object and touching the unevenness of the object is higher than the frequency at which the skin vibrates by pushing the finger into the object.
- Merkel cells 2004 having a sensitivity peak in a low frequency range 2006 such as approximately 10 Hz or less are mainly ignited for the pushing operation.
- the Pachiny body 2003 having a sensitivity peak in a high frequency range 2007 of about 100 Hz and the Meissner body 2005 having a sensitivity peak in a middle frequency 2008 of about 30 Hz mainly ignite. It will be.
- a person can switch the frequency of the horizontal axis 2002 in FIG. 8 by changing the action of touching an object, that is, the “hand movement”.
- vibration stimulation is applied to a plurality of tactile receptors.
- the softness and softness feeling 1804 and the frictional feeling 1805 in FIG. 5 are not sensed at the same time but sensed by time division. That is, the softness and softness 1804 is a tactile sensation that is sensed by a pushing operation, and the Merkel cell 2004 is primarily ignited and sensed in a low frequency range 2006 in terms of vibration frequency. In this pushing operation, since the pressing direction is the normal direction of the object, it is unlikely that vibrations in the high frequency range 2007, which are mainly caused by collisions with surface irregularities, occur. As long as the vibration stimulus is not input to the tactile receptor, the tactile sensation that each tactile receptor bears does not occur, and in the push-in operation, a softness and softness 1804 is sensed centering on the Merkel cell 2004.
- tactile sensation feature vector 1602 As described above, people have different tactile sensations by changing the movement of the hand touching the object, that is, the “tactile movement”. On the other hand, if all of the plurality of tactile sensations felt from the object are collectively represented as a tactile sensation feature vector 1602, there arises a problem that the state that “a person feels various tactile sensations individually in time” cannot be represented. . For this reason, when a material search apparatus that searches for materials having similar tactile sensations using the tactile sensation feature vector 1602 is assumed, it is not possible to perform a search that focuses on tactile sensations that people feel different for each tactile movement.
- the tactile sensation processing device 2100 illustrated in FIG. 6 has a problem of excessive operation in which a tactile sensation that is not felt by a person is always presented. That is, since the plurality of tactile sensations felt from the test object 1709 are collectively represented as the estimated tactile sensation feature vector F ′, the tactile sensation display unit 1801 presents a plurality of tactile sensations to the person 1802 at the same time. Become.
- the actuator unit 2102 presents a feeling of roughness 1803 and a feeling of friction 1805 at the same time, but due to the hand movement described in FIG. 7 and the vibration frequency selectivity of the tactile receptor described in FIG. Does not feel the feeling of roughness 1803 and the feeling of friction 1805 at the same time while feeling the hardness 1804.
- the present invention has been made to solve the above-described problems, and its purpose is to make it possible to sense and reproduce various tactile sensations that humans feel individually in time by time division. Is to provide.
- a tactile sensation processing apparatus includes a motion measurement unit that measures a contact state between a person and an object, a physical measurement unit that measures a physical characteristic value of the object, and a person touching an arbitrary object. Based on the physical property value according to the contact state and the feature amount related to the tactile sensation when the arbitrary object is touched, based on the physical property weight of the arbitrary object for each contact state, A physical psychological conversion unit that generates a feature value related to tactile sensation from the physical characteristic value measured by the physical measurement unit.
- the tactile sensation processing apparatus may further include an output unit that outputs a feature amount related to the tactile sensation generated by the physical psychological conversion unit as a tactile sensation felt when the person touches the object.
- the physical measurement unit can measure physical characteristic values related to a plurality of types of physical characteristics of the object, and the physical measurement unit can measure physical characteristics measured according to a contact state where the person touches the object. You may change the type.
- the movement measurement unit measures a feature amount related to a hand movement when the human hand is in contact with the object, and the feature amount related to the hand movement includes a position coordinate of the hand, the hand and the hand At least one of a contact position coordinate with an object, a change amount of the position coordinate, a moving speed of the hand, and a moving acceleration of the hand may be included.
- the physical measurement unit may measure at least one of the shape and stress of the object as a physical characteristic value of the object.
- the feature amount related to the tactile sensation may include at least one of a concavo-convex sensation, a hardness / softness sensation, a friction sensation, and a viscosity sensation defined as factors extracted from the sensory response of the subject.
- the physical psychological conversion unit holds information that associates a physical characteristic value of the object created in advance and a feature amount related to a tactile sensation when the object is touched, and the information includes the information
- a table associating feature quantities relating to hand movements, physical characteristic values of the object and feature quantities relating to the tactile sensation, or input of feature quantities relating to hand movements and physical characteristic values of the learning object It may be a function that outputs a feature amount related to tactile sensation.
- the tactile sensation processing apparatus further includes a tactile sensation display unit that receives a feature amount related to the tactile sensation from the output unit and generates a force based on the received feature amount related to the tactile sensation, thereby causing the tactile sensation to occur.
- the tactile sensation display unit includes a psychophysical conversion unit that converts a feature quantity related to the tactile sensation received from the recording unit into a control signal based on a conversion rule prepared in advance, and a toucher's hand based on the control signal.
- a driving unit that applies force to the tactile sensation display unit, and a hand movement measuring unit that measures a feature amount related to the movement of the toucher's hand that has touched the tactile display unit. You may receive the feature-value regarding the said tactile sense recorded on the said recording part specified based on a movement feature-value.
- Another tactile sensation processing device includes a tactile sensation display unit that generates a tactile sensation by generating a force, a physical measurement unit that measures a physical property value of a test object given in advance, Based on the prepared physical measurement value of the learning object and the information that associates the feature value related to the tactile sensation when the person touches the learning object, the measured physical property value of the test object
- the tactile sensation display unit converts the feature amount related to the tactile sensation into a control signal based on a conversion rule prepared in advance, and the control signal.
- a driving unit that applies a force to the toucher's hand based on the touch sensor, and a hand movement measuring unit that measures a feature quantity related to the movement of the toucher's hand that has touched the tactile display unit. Parts, based on the feature amount relating to movement of the hand that is measured, to measure the physical property values of the test object.
- Still another tactile sensation processing device includes a physical measurement unit that measures a physical characteristic value of a test object using a physical sensor, and a physical measurement value of the physical sensor when measuring the physical characteristic value of the test object.
- Physical sensor movement instruction unit for instructing movement, physical characteristic value of the learning object, feature amount relating to hand movement when the subject touches the learning object, and feature amount relating to tactile sensation when the object is touched
- the measured physical property value is based on information associated in advance so that the type and weight of the physical property value referred to according to the feature value related to the hand movement is changed.
- a psychophysical conversion unit that converts the characteristic quantity relating to the identified tactile sensation as an amount of tactile sensation that the person feels when touching the test object. .
- the tactile sensation processing apparatus further includes a tactile sensation display unit that receives a feature amount related to the tactile sensation from the output unit and generates a force based on the received feature amount related to the tactile sensation, thereby causing the tactile sensation to occur.
- the tactile sensation display unit includes a psychophysical conversion unit that converts a feature quantity related to the tactile sensation received from the recording unit into a control signal based on a conversion rule prepared in advance, and a toucher's hand based on the control signal.
- the psychophysical conversion unit comprising: an actuator unit that applies a force to the toucher and causes a tactile sensation to the toucher; and a hand motion measurement unit that measures a feature amount related to the toucher's hand touching the actuator unit. May receive a feature quantity related to the tactile sensation recorded in the recording unit, which is specified based on a movement feature quantity of the toucher's hand.
- tactile sensation sensing and tactile sensation presentation according to the contact state between a human hand and an object can be performed, so that only tactile sensation information necessary for human sensation can be handled without waste.
- the actuator of the tactile display needs to drive only the part necessary for human sense, and the configuration and control of the actuator can be simplified.
- FIG. 6 is a diagram showing a tactile sensation feature value vector 1602 described in a four-dimensional feature value space 1601. It is the block diagram which showed the flow of the processing of the prior art of tactile quantification and tactile sensing. It is a figure which shows the relationship between the structure of the tactile sensation display part 1801, and the feeling of roughness 1803 which is a tactile sensation of the person 1802, the softness feeling 1804, and the friction feeling 1805.
- FIG. 6 shows the structure of the tactile sensation display part 1801, and the feeling of roughness 1803 which is a tactile sensation of the person 1802, the softness feeling 1804, and the friction feeling 1805.
- FIG. 6 is a block diagram of a tactile sensation processing device 2100 configured by the inventor of the present application virtually combining the tactile quantification and tactile sensing described in FIG. 4 and the tactile display described in FIG. 5.
- (A) is a diagram showing a tracing operation 1902 of a finger 1901
- (b) is a diagram showing a pushing operation 1901.
- This is information having sensitivity distributions of Merkel cells, Meissner bodies, and Patini bodies. It is a figure which shows the position and length 103 of the starting point 102 of the applied pressure which a human hand applied. It is a figure which shows the structural example of the conventional raw material search apparatus. It is a figure which shows the example of the conventional material search method which designates the kind of tactile sensation, and the tactile sensation intensity
- FIG. 1 It is a figure which shows the example of a result display of the search process shown in FIG. It is a figure which shows the western dressing detailed data regarding the material number 53 among the search results of FIG. It is a figure which shows the structure mainly utilized when the raw material search apparatus 800 based on Embodiment 1 performs a search. It is a figure which shows the example of arrangement
- (A) And (b) is a top view and A-A 'cross section of the sensor. 5 is a diagram illustrating a configuration example of an object physical measurement unit 109.
- FIG. (A) And (b) is a figure explaining tactile-motion measurement using an image sensor. It is a figure which shows the example which attached the marker to the fingertip or the joint.
- FIG 3 is a diagram in which a piezoelectric sensor 401 is installed under a test object 105. It is a figure which shows an example of the data structure of tactile movement DB13. It is a conceptual diagram which shows the structure of tactile weight DB18. It is a figure which shows an example of the data structure of material DB23. It is a figure which shows the structure mainly utilized when the raw material search apparatus 800 based on Embodiment 1 performs learning. It is a figure which shows the structure mainly utilized when the material search apparatus 800 based on the tactile sense concerning Embodiment 1 performs learning, and constructed
- Tactile operation feature quantity vector H of the human 107 is a diagram for explaining a method of selecting the H 'by using, Sawado estimated tactile feature quantity vector F presents the human 107'.
- (A) And (b) is a figure which shows the tactile movement feature-value vector space.
- FIG. 10 is a diagram showing that a tactile motion feature quantity vector 901 outside an area 601 (person 104) is mapped to the nearest location in the area 601 and a tactile action feature quantity vector 903 is detected in a form having an error 902. . It is a block diagram which shows the structure of the tactile sensation processing apparatus 1000 by Embodiment 5.
- FIG. 10 is a block diagram illustrating a tactile sensation processing device 1100 according to a sixth embodiment.
- FIG. It is a figure which shows the example of the XYZ stage.
- A is the figure which defined the area
- (b) is that the vector group distributed equally in the area
- FIG. It is a block diagram which shows the tactile sensation processing apparatus 1400 by Embodiment 7.
- the present invention pays attention to the fact that “people feel various tactile sensations individually in time”, and uses this.
- the source of tactile sensation is in the firing of tactile receptors present on human fingers, the source of the tactile receptor firing is in skin vibrations, and the source of skin vibrations is in human hands. Occurs when the object contacts the object.
- the present invention quantitatively describes the tactile sensation felt by a person based on the contact state between the person's hand and the object.
- the contact state between a human hand and a stationary object is determined by the movement of the human hand. That is, when the object does not move, the contact state can be expressed by what position the human hand touches the object and what force is applied.
- the pressing force applied by the human hand can be expressed by the position and length 103 of the starting point 102 of the pressing force vector 101, and the starting point 102 of the pressing force vector 101 is the finger 1901 of the person 1802 and the object. 1903 contacts.
- the stimulation that the human skin elastic body receives is determined.
- the physical characteristics of the object 1903 are, for example, hardness, elasticity, viscosity, surface shape, and the like, and means for measuring these is required.
- any physical property value may be measured excessively. However, it is basically desirable to measure only the physical characteristic values related to the pressurization of the human hand with less waste. For example, when a human hand pushes in, the hardness, normal elasticity and viscosity of the object are measured, and the surface shape and tangential elasticity and viscosity are not measured. On the other hand, when a human hand performs a tracing operation, it is preferable to measure the surface shape, tangential elasticity and viscosity, and not to measure the hardness, normal elasticity and viscosity of the object.
- the present invention uses this process as a black box, directly associates physical characteristic values measured in advance with the learning object and tactile sensation amounts, and performs conversion from physical characteristic values to tactile sensation amounts by matrix conversion, table reference, etc. To do.
- FIG. 1 is a block diagram illustrating the principle of the tactile sensor in consideration of the above-described tactile movement, which is divided into a learning process 2305 process and an after-learning execution process 2306 process.
- FIG. 1 shows a configuration in which hand movement measuring means 2301 is added to FIG. 4 described as an example of the prior art.
- factor analysis means 2302, physical psychological conversion calculation means 2303, and physical psychological conversion means 2304 are different from those in FIG. 4 of the prior art.
- the learning step 2305 the measured tactile movement is classified into a plurality of patterns as tactile movement feature quantity vectors, factor analysis is performed for each pattern individually, and the learning material physical characteristic value vector P S is used as the tactile feature quantity.
- a function M H to be converted into the vector F H is calculated individually for each tactile movement pattern.
- the execution step 2306 to select the function M H in accordance with the operation tactile touch test object, converts the reference material physical property value vector P t to the tactile feature quantity vector F 'H.
- the tactile display that reproduces the tactile sensation measures the tactile movement of the person 2307 touching the actuator. Based on the similarity of the measured tactile motion feature vector, a tactile feature to be reproduced on the tactile display is selected, and the tactile display is driven to present the tactile feature.
- the selected tactile sensation feature amount can be referred to as an estimated value of the tactile sensation amount.
- the tactile sensation is a factor extracted from a human sensory response, and a psychological quantity (for example, human response to a question as shown in FIG. 2) relating to tactile sensation (for example, at least one of unevenness, hardness, friction, and viscosity). Expressed as quantity).
- Embodiments 1 to 3 As embodiments in which the tactile sensation estimated according to the movement of the human hand is used, two types of embodiments are shown below.
- One is related to a material retrieval apparatus that retrieves a material similar to this based on the tactile sensation of a material that a person feels when touching the material, and is disclosed in Embodiments 1 to 3 below.
- the other relates to a tactile display that reproduces and presents the tactile sensation of a material, and is disclosed in Embodiments 4 to 7.
- FIG. 10 shows a block diagram of the above-described conventional material search apparatus.
- a conventional material search apparatus will be described with reference to an example of a search screen.
- FIG. 11 is a search screen of a conventional material search apparatus.
- the user can select tactile vocabulary such as “koshi”, “hari”, and “squeak” indicating tactile sensation as a search item, and the tactile strength of each item is “Yes”, “Medium”, Specify in three stages, “No”.
- the material search apparatus accepts these as search keys.
- “search” is selected as the search item, and “Yes” is selected as the degree.
- FIG. 12 is an example in which 130 relevant materials are extracted from the search key input in FIG. 11 and 20 pieces of information are displayed.
- a keyword is used as a search key. For example, “hemp” as the material type, “plain weave” as the weaving type, and “40” representing the thickness of the thread can be further narrowed down as keywords.
- FIG. 13 shows detailed Western dress data regarding the material number 53 in the search result of FIG.
- the fabric design data describes the color of the fabric, the thickness weaving width, the type of warp and weft fibers used in the fabric, the thickness, the twisting method, and the like.
- the measured data includes physical quantities such as crease wear actually measured and tactile strength for tactile vocabulary such as “strain”.
- the operation of the material search apparatus of FIG. 10 is divided into learning and retrieval.
- the material search device creates a tactile sensation database (DB) 503 and a fabric design database (DB) 505.
- the material search unit 507 searches the tactile sensation database 503 using the tactile vocabulary and its strength as search conditions, and creates a corresponding base list.
- the search result display unit 508 displays the western dressing detailed data with reference to the fabric design database 505 for the material selected from the material list of the search result according to the instruction from the searcher.
- Equation 1 Relationship of the material physical property values xi and tactile strength H k of each tactile vocabulary Y k shown in Equation 1 is a row vector C k to the weighting coefficients C k0, C ki and components, and the component material physical property value xi
- the column vector X to be used can be described as follows.
- the row vector C k is referred to as “tactile weight vector”
- the column vector X is referred to as “material physical characteristic value X”.
- the estimated tactile sensation strength calculation unit 502 calculates the tactile sensation strength H k for all tactile sensations from the material physical property value X measured by the material physical quantity measurement unit 501 using Equation 7.
- the estimated tactile sensation strength H k calculated by the estimated tactile sensation strength calculating unit 502 is stored in the tactile sensation database 503 in association with the material number for identifying the material.
- the fabric design data acquisition unit 504 acquires information such as the thickness of the yarn constituting the material fabric, the twisting method, and the weaving method of the fabric, and stores the information in the fabric design database 505 in association with the material number for identifying the material.
- the search key input unit 506 inputs the search key of the conventional material search apparatus described in Non-Patent Document 2.
- the user has tactile vocabulary such as “koshi”, “hari”, and “squeak” indicating tactile sensation. It can be selected as a search item, and for each item, the tactile sensation strength can be specified in three levels: “Yes”, “Medium”, and “No”.
- the material search apparatus accepts these as search keys.
- the search keyword input unit 506b accepts a keyword as a search key, and is used to further narrow down the search result (FIG. 12) based on the tactile vocabulary and the tactile sensation strength by the keyword.
- the material search unit 507 searches the tactile sensation DB 503 for a material that matches the search conditions input by the search item and the degree selection unit 506a, and outputs a material number list of the material.
- the search result display unit 508 displays search results from this list in a display mode as shown in FIG. 12, for example.
- the searcher further selects a desired material from the list, thereby searching for detailed Western clothing detailed data as shown in FIG. 13 from the fabric design database 505 and displaying it.
- Such a material retrieval apparatus has the following problems.
- the tactile vocabulary used as a search condition shown in FIG. 11 is a term of an industry expert, and it is difficult for general consumers to imagine the tactile sensation. Moreover, even if the tactile vocabulary is rewritten to words that can be understood by general consumers, such as “soft”, the specific tactile sensation image varies from person to person. Similarly, even if the intensity of tactile sensation is expressed as a certain, medium, no, or numerical value, the image of the specific intensity varies from person to person. For this reason, it may happen that the result of the material search under such linguistic and numerical search conditions is completely different from what the searcher imagined.
- a material retrieval apparatus that solves such a problem is shown in the following first to third embodiments.
- tactile sensation can be expressed by “kind of tactile sensation” and “tactile sensation strength”. For example, a material whose “feel of roughness” can be felt at a tactile strength of 5 as the type of tactile sensation. However, when the apparatus of the present invention is used, it is not necessary for the user to input the word “roughness” indicating the type of tactile sensation and the numerical value “5” indicating the tactile sensation strength. “Type of tactile sensation” and “tactile sensation strength” are criteria for the material retrieval apparatus according to the present invention to classify materials qualitatively and quantitatively.
- the material search device acquires parameters indicating “kind of touch” and “tactile strength” from the movement of the user's hand, and searches the database based on these parameters.
- the tactile sensation strength H is estimated by the equation 8 obtained by correcting the equation 7.
- m is a variable representing the tactile movement, and is hereinafter referred to as a tactile movement variable m.
- This expression means that the tactile weight vector for obtaining the tactile sensation intensity changes depending on the tactile movement variable m, and different tactile intensity H can be obtained if different tactile movements are performed on a certain material. .
- the tactile strength H (m1) when the tracing operation is performed on the material is different from the tactile strength H (m2) when the pushing operation is performed on the material.
- the type of tactile sensation that is noticeable varies depending on the tactile movement, and it is considered that when the user performs a tracing action, the material feels uneven.
- the operation of pushing the material is performed, it is considered that the softness of the material is strongly felt.
- the tactile sensation of the person is estimated by obtaining a relationship between the tactile movement and the tactile sensation in advance by conducting a tactile experiment using the subject. Is possible. Therefore, learning through the subject experiment eliminates the need to express the tactile sensation that people are feeling in words or numerically, and eliminates the instability of verbalization and numericalization. Whether you are an expert who has a deep understanding of the tactile vocabulary or a general user who is not familiar with the tactile vocabulary, you should get the same tactile sensation if the tactile receptor firing level is the same. It can be said that the tactile sensation can be estimated regardless of the toucher by measuring the tactile motion that determines the vibration given to the touch.
- the type of tactile sensation that is strongly felt As described above, when the tactile movement changes, the type of tactile sensation that strongly feels also changes. In other words, when a person wants to feel a certain kind of tactile sensation strongly, he can say that he / she performs the most suitable tactile movement. Therefore, if the learning about the types of tactile movements and tactile sensations is performed in advance, it is possible to estimate what kind of tactile sensation the toucher is going to feel by observing the tactile movements. In the conventional example shown in Non-Patent Document 2, the type of tactile sensation is defined as shown in FIG. 11, and material search is performed under the condition of how much the tactile sensation strength is. A function similar to this can be realized without estimating the tactile sensation that the toucher feels strongly from the tactile movement, without the process of verbalizing the tactile sensation as in the conventional example.
- FIG. 14 is a configuration diagram of the material search device 800 in the present embodiment.
- the toucher 20 uses the material search apparatus 800 in order to find a material having a specific tactile sensation using a search server on the network.
- the toucher 20 touches a reference material having a tactile sensation to be searched. For example, when the user wants to search for a material with a soft feeling felt when the reference material 22 is pushed in, the toucher 20 pushes the reference material 22 into the material. When the user wants to search for a material with a smooth feeling when tracing the surface of the material, the toucher 20 traces the surface of the reference material 22.
- the material retrieval apparatus 800 measures the tactile motion at that time and the physical property value of the reference material, and estimates the type of tactile sensation felt by the toucher and the tactile sensation strength from these.
- the rough processing flow of the material search apparatus 800 is as follows.
- the material physical quantity measurement unit 1 measures the material physical characteristic value X of the reference material 22, and the motion physical quantity measurement unit 10 measures the movement of the finger touching the reference material 22 by the toucher 20.
- the measured tactile movement of the toucher 20 is a tactile movement variable m which is a variable for finally identifying a finite number of tactile movement types by the movement feature quantity calculating unit 11, the tactile movement DB 13, and the tactile movement estimating unit 17. Is converted to Since the tactile motion variable m is regarded as a motion for feeling a certain tactile sensation, the tactile weight vector search unit 19 uses the tactile weight for estimating the tactile strength of this tactile sensation from the tactile motion variable m. A vector C (m) is determined.
- the material tactile sensation estimation unit 2 uses the tactile weight vector C (m) determined based on the material physical property value X of the reference material and the tactile movement, and the toucher 20 touches the reference material 22 according to the above equation (8).
- the tactile sensation strength H corresponding to the tactile sensation motion variable m is estimated.
- the tactile sensation intensities of a plurality of search target materials are stored in advance for each tactile motion variable m (that is, for each tactile sensation type).
- the material search unit 3 searches the material DB 23 for a material closest to the tactile intensity H calculated by the material tactile sensation estimation unit 2 using the tactile movement variable m of the toucher 20 as a condition. The result is displayed on the search result display unit 7.
- the toucher 20 can check the search result by text information, image information, or the like displayed on the search result display unit 7. Thus, a material having a tactile sensation similar to the reference material is retrieved and presented with respect to the type of tactile sensation estimated from the tactile movement of the toucher 20.
- the material physical quantity measuring unit 1 measures the material physical characteristic value X of the reference material. It is arbitrary what kind of measurement value the material physical characteristic value X is captured. For example, a surface shape corresponding to a feeling of flatness / concaveness, a friction characteristic corresponding to a feeling of smooth / roughness, an elastic amount corresponding to a feeling of softness / hardness, and the like. Moreover, the measuring method of the material physical property value X is also arbitrary.
- the sensor 201 in FIG. 15A measures the material physical characteristic value X.
- the sensor 201 is fixed to the finger so as to come into contact with the material in accordance with the movement of the hand or the finger, and simultaneously measures the material physical property value X in the operation in which the user feels the touch of the material.
- FIGS. 16A and 16B are top views of a sensor capable of simultaneously measuring a surface shape corresponding to a flat / concave feeling, a friction characteristic corresponding to a smooth / rough feeling, and an elasticity corresponding to a soft / hard feeling. It is a figure and AA 'sectional drawing. This sensor is introduced in Non-Patent Document 8, for example.
- This sensor is shaped like a fingertip. As shown in FIG. 16A, in this sensor, five strain gauges 201a are embedded in the elastic body along the X axis and two along the Y axis. As shown in FIG. 16B, force sensors 201b are arranged on the x-axis and the z-axis.
- Characteristic quantities corresponding to tactile movements are the material physical characteristic value strain gauge 201a and force. It is calculated using the output of the sensor 201b. Specifically, the spatial frequency and the amplitude of the unevenness are calculated based on the strain change of the strain gauge 201a, and the direction of the unevenness is calculated based on the output ratio of the strain gauge 201a. These are feature amounts corresponding to the surface shape corresponding to the flatness / unevenness feeling. Furthermore, the variance of the strain distribution of the strain gauge 201a is also calculated. This corresponds to a friction characteristic corresponding to a smooth / rough feeling. Then, by obtaining the ratio of the force in the normal direction and the tangential direction of the force sensor 201b, an elastic amount corresponding to the softness / hardness feeling can be obtained.
- FIG. 17 shows a configuration example for measuring a physical characteristic value of a test object by moving a physical measurement sensor on an XYZ three-dimensional stage.
- the XYZ three-dimensional stage 2501 holds the physical measurement sensor 2504 with two support poles 2502 and 2503.
- the support pole 2502 can be moved by the motor 2505 along the rail 2506 in the X-axis direction.
- the motor 2507 can be moved along the rail 2508 in the Z-axis direction.
- the support pole 2503 can be moved in the Y-axis direction along the rail 510 by the motor 2509.
- the motor control unit 511 controls the movements of the motor 2505, the motor 2507, and the motor 2509 according to the tactile movement feature quantity vector H, and moves the physical measurement sensor 2504 with the same movement as the hand of the toucher 20. That is, when the toucher 20 performs a pushing operation, the motor 2507 is driven to move the physical measurement sensor 2504 in the Z-axis direction. On the other hand, when the toucher 20 performs the tracing operation, the motor 2505 and the motor 2509 are driven to move the physical measurement sensor 2504 on the XY plane.
- the present invention does not limit the type of the physical measurement sensor 2504, and any physical measurement sensor can be used.
- any physical measurement sensor can be used.
- the tactile sensation felt by a person is estimated from the physical characteristic value, the accuracy of the estimation satisfies a predetermined standard. It is necessary to perform physical measurement to satisfy the condition. Therefore, in many cases, such as a hardness meter that measures the hardness of an object and a friction meter that measures the coefficient of friction, there are a plurality of physical property values to be measured, and a single function that can measure one physical property value.
- the physical measurement sensor 2504 is replaced and measurement is performed in a plurality of times.
- the physical measurement sensor 2504 of the present invention is a sensor that performs measurement without contact with an object, as shown by the fact that physical characteristics such as surface shape can also be measured by optical measurement means such as a laser. It does not matter.
- FIG. 18 is a diagram for explaining tactile movement measurement using an image sensor.
- FIG. 18A shows a state where the hand 2201 is touching the test object 105.
- the image sensor 202 captures the entire hand 2201 and shoots the hand 2201 continuously in time by continuous shooting of video moving images or still images.
- the image feature point extraction unit 203 extracts edges and points having a large luminance change for each captured image.
- the hand movement feature calculation unit 204 calculates the touch action feature vector H using the pixel values constituting the edges and points extracted by the image feature point extraction unit 203. Specifically, as shown in FIG. 18B, two images of time T1 and time T2 that are temporally continuous, the image 205 and the image 206 are overlapped, and the feature point having the same pixel value is tracked. Thus, the change in the x-coordinate and the y-coordinate is set as the touch action feature amount.
- the thumb coordinate change (x1 y1) is the thumb motion vector 207
- the index finger coordinate change (x2 y2) is the index finger motion vector 208
- the ring finger coordinate change (x3 y3) is the ring finger motion vector 209.
- the combined motion vector (x1, y1, x2, y2, x3, y3) is defined as a touch motion feature vector H.
- the number of dimensions of the tactile motion vector is arbitrary. If two image sensors are used, the distance from the image sensor to the feature point can be calculated based on the principle of triangulation, and the distance D is added to the xy coordinates to form a three-dimensional vector. Become. Further, the contact position of the object and the hand itself is important information, and not only the coordinate change but also the hand position coordinate (x, y) or (x, y, z) is also important information.
- FIG. 19 shows an example in which markers are attached to fingertips and joints.
- markers are attached to the fingertips, ten to the finger joints, and one to the wrist.
- the image sensor 202 captures the hand 2201 including the marker 301, and the marker extraction unit 302 extracts the marker 301 from the captured image.
- the brightness or color of the marker 301 is made different from that of the hand 2201 or the test object 105 so that the marker 301 can be accurately extracted.
- the hand movement feature calculation unit 204 tracks the coordinates of the marker center from time T1 to time T2 in the same manner as in FIG. 18B, and uses the change in the x and y coordinates as the tactile movement feature quantity vector H. .
- the acceleration sensor When performing the tactile movement measurement with the acceleration sensor, the acceleration sensor is attached to the fingertip, the joint or the like in the same manner as the marker attachment of FIG.
- an acceleration sensor unlike the image sensor 202, it is not necessary to shoot separately, and the mounted acceleration sensor itself performs measurement.
- the velocity is obtained by differentiating the acceleration
- the position is obtained by second-order differentiation
- the movement of the corresponding feature point is calculated from the position information in the same manner as in the case of the image sensor.
- a quantity vector H is calculated.
- the piezoelectric sensor 401 When the tactile movement measurement is performed by the piezoelectric sensor, the piezoelectric sensor 401 is installed under the test object 105 as shown in FIG.
- the piezoelectric sensor 401 has piezoelectric elements that generate electricity by pressurization in a two-dimensional array, and can measure the position where the hand 2201 is in contact with the test object 105.
- the motion feature quantity calculation unit 11 converts the motion physical quantity measured by the motion physical quantity measurement unit 10 into a tactile motion feature quantity vector v which is a time series of vectors characteristically representing the motion. This is used when the tactile movement estimation unit 17 determines the type of tactile movement. Specifically, for example, the XYZ coordinates of the hand / finger at each time t, the speed in the XYZ axis direction, the rotation, and the like are recorded along with the time change. Alternatively, the motion can be expressed for each of the XY-axis rotation component, the YZ-axis rotation component, the ZX-axis rotation component, or a total of six components.
- the repetitive motion is detected by a method such as autocorrelation analysis, and the above parameters for one cycle and the parameters indicating the number of repetitions and the cycle are determined. It may be a combination. Furthermore, each parameter may be data quantized to a resolution with no practical problem.
- the tactile movement estimation unit 17 searches the tactile movement DB 13 from the tactile movement feature quantity vector v that is an output from the movement feature quantity calculation unit 11 and outputs a tactile movement variable m.
- the touch motion DB 13 describes the relationship between the touch motion feature vector v and the touch motion variable m that identifies the type of touch motion.
- the tactile movement may have an infinite variety of movements, considering a slight difference in speed. However, if it is limited to the purpose of estimating the kind of tactile sensation and the tactile sensation strength, it is considered that tactile movements are grouped into finite n groups (grouping is possible). Each of these groups corresponds to, for example, “race operation”, “push-in operation”, and the like.
- the tactile movement variable m is a variable for identifying which group of tactile movements.
- the tactile action variable m may be a simple identification label indicating a group, or may be a tactile action feature vector v ave of an average action of the group.
- What is stored in the touch motion DB 13 is the relationship between the touch motion feature vector v corresponding to various touch motions and the touch motion variable m indicating the group to which the touch motion feature vector v belongs.
- a look-up table from the touch motion feature vector v to the touch motion variable m can be used.
- FIG. 21 shows an example of the touch action DB 13 in the look-up table format.
- the tactile movement feature quantity vector v is assumed to be a time-series pattern of the moving speed in each XYZ axis direction and the rotational angular velocity of each XYZ axis.
- an average tactile movement pattern for each grouped tactile movement is expressed by using a tactile movement feature quantity vector v.
- the Y-axis direction is expressed. A pattern in which only reciprocating motion is repeated is described.
- the touch motion estimation unit 17 searches the touch motion DB 13 using the touch motion feature vector v obtained from the touch motion of the toucher 20 to obtain a touch motion variable m. In this search, a similarity comparison based on the inter-vector distance between the touch motion feature vector v of the toucher 20 and the touch motion feature vector vi for each touch motion variable mi in the touch motion DB 13 is performed, and the most similar touch Selects and outputs the operation variable mi. As illustrated in FIG. 21, the tactile movement is generally described as a time-series pattern with repetition. It is considered that the tactile motion feature vector v obtained from the touch motion of the toucher 20 and the touch motion feature vector stored in the touch motion DB 13 are different in vector time series length.
- a collation technique such as Dynamic Time Warping that can collate by non-linear compression / decompression of time series patterns.
- a known method can be used as a technique for classifying operations in advance and determining the type of operation from an operation signal obtained from a sensor.
- a method of determining the entire action type based on a series of labels assigned to partial actions is also possible.
- the tactile weight vector search unit 19 searches the tactile weight DB 18 to obtain a tactile weight vector C (m). That is, the tactile weight DB 18 is a database in which information on the correspondence between the material physical property value X and the tactile intensity H based on Equation 8 is recorded in association with the tactile movement, and the tactile movement variable m and the tactile weight vector C ( The relationship of m) is stored.
- the tactile movement is regarded as an action for obtaining a tactile sensation of one type of tactile sensation. Therefore, the tactile movement variable m corresponds to one tactile sensation type.
- the tactile weight C (m) in the tactile weight DB 18 is a tactile weight vector for obtaining the tactile intensity of the tactile type corresponding to the tactile motion variable m.
- the format of the tactile weight DB 18 is most simply a lookup table from the tactile motion variable m to the tactile weight row example C (m).
- FIG. 22 shows the concept of the tactile weight DB 18 expressed in the format of Table E01.
- the tactile weight DB 18 has a tactile weight vector C (m) for each tactile motion variable m as shown in FIG.
- the tactile weight vector C (m1) is selected and used as the tactile weight vector C (m) of Equation 8.
- the tactile weight vector C (m) needs to be learned in advance before executing the search. Details of this will be described later.
- the material tactile sensation estimation unit 2 uses the material physical characteristic value X of the reference material measured by the material physical quantity measurement unit 1 and the tactile weight vector C (m) determined based on the tactile motion, and calculates the tactile sensation according to the above equation (8).
- the tactile sensation strength H that the person 20 feels by touching the reference material 22 is estimated. As described above, this tactile sensation strength is a tactile sensation strength related to the type of tactile sensation corresponding to the tactile movement variable m.
- the material DB search unit 3 uses the tactile motion variable m of the toucher 20 and the tactile strength H calculated by the material tactile sensation estimation unit 2, and uses a similar tactile material from among the search target materials stored in the material DB 23. Search for. In the search, the difference between the tactile sensation intensity H of the material tactile sensation estimation unit 2 and the tactile sensation intensity H ′ in the same tactile movement variable m of the search target material is used as a search score. A material having the best score (minimum distance difference) or a material group whose score satisfies a predetermined condition (distance difference is within a certain range) is output as search results together with the search score information.
- the material DB 23 describes the relationship among materials, tactile movements, and tactile sensations.
- the type of tactile sensation is uniquely determined from the tactile movement variable m and is not explicitly required, but is described for convenience.
- the tactile action defines the type of tactile sensation felt by the toucher, and at the same time affects the tactile intensity of the tactile sensation.
- the search result display unit 7 receives the result of the material DB search unit 3 and presents this result to the user searching for the material.
- the content is arbitrary, such as the name of the material, the product name, or a picture of the material, a video that captures the situation when the material is applied, and other physical property values related to the material.
- an index indicating any material in the sample material group held by the searcher may be used.
- the searcher can confirm the search result by touching the sample material.
- you may use the tactile sensation presentation apparatus which reproduces the tactile sensation of a raw material. For example, for presentation of a rough / smooth feeling, for example, a tactile sense presentation device using an electrostatic actuator described in Patent Document 2 can be used.
- a tactile sensation presentation device by pin movement described in Patent Document 3 can be used.
- a tactile sense presentation device using ultrasonic vibration described in Patent Document 4 may be used.
- the number of results displayed is not limited to one material that is closest to the estimated tactile sensation, but may be a plurality of materials that are similar within a certain range. At this time, it is also effective that the order of the presented materials is rearranged based on the search score information added to the search result.
- the flow of search execution so far is a method of searching for a material by estimating the type of tactile sensation and the strength of tactile sensation from a group (a series) of tactile operations performed on the reference material 22 assuming one type of tactile sensation. Met. Furthermore, it is more effective if the final output is obtained by performing an AND operation or an OR operation on the search results of each tactile movement from tactile movements performed in multiple times. In this case, it is possible to search for a material desired by the searcher by changing the type of the touch operation or changing the reference material for each touch operation. For this purpose, a combination instruction unit for instructing an AND / OR operation may be provided.
- each component shown in FIG. 14 is realized by the following hardware. That is, the material physical quantity measuring unit 1 and the motion physical quantity measuring unit 10 can be realized by using a sensor as in the above-described example.
- the motion feature amount calculation unit 11, the tactile motion estimation unit 17, the tactile weight vector search unit 19, the material tactile sensation estimation unit 2, and the material DB search unit 3 can be realized using, for example, a computer.
- the tactile movement DB 13, the tactile sensation weight DB 18, and the material DB 23 can be realized by a recording medium such as a hard disk drive (HDD), a memory card, or a solid state drive (SSD).
- HDD hard disk drive
- SSD solid state drive
- the material search execution method by the material search device 800 has been described above. In the following, learning steps such as construction of each database necessary for material search execution will be described with reference to FIGS.
- the learning step 1 for constructing the tactile motion DB 13 and the tactile weight DB 18 shown in FIG. 24 is a step that may be performed once when the material search apparatus 800 is constructed. However, each database may be reconstructed as necessary.
- the learning step 2 for constructing the material DB 23 shown in FIG. 25 is a step that must be performed every time a search target material is added after the tactile weight DB 18 is constructed.
- each step will be described step by step.
- the learning step 1 for constructing the tactile motion DB 13 and the tactile weight DB 18 shown in FIG. 24 will be described.
- tactile sensation intensities are collected for a plurality of tactile sensations when a plurality of touchers touch a plurality of learning materials.
- the tactile movements of the toucher at this time are also collected, and these are statistically collected to construct the above two databases.
- each toucher 20 touches each learning material 21 based on each tactile vocabulary presented by the tactile vocabulary specifying unit 9 in order to define the type of tactile sensation.
- the tactile sensation strength felt by the toucher 20 at this time with respect to the designated vocabulary is recorded by the tactile sensation strength recording unit 14.
- the tactile movement of the toucher 20 at that time is acquired by the movement physical quantity measuring unit 10, the tactile movement feature quantity vector v is obtained by the movement feature quantity calculating section 11, and recorded together with the tactile vocabulary.
- the material physical characteristic value X of each learning material 21 is measured by the material physical quantity measuring unit 1 in advance.
- the tactile movement learning unit 12 collects all collected tactile movement feature quantity vectors v, and classifies them into similar groups of haptic movement feature quantity vectors v. Then, the tactile movement variable m for identifying each group is associated with the average tactile movement feature quantity vector v ave of the group and stored in the tactile movement DB 13.
- the tactile sensation weight calculation unit 15 uses the tactile strength data that is different for each of the toucher, the learning material, and the tactile sensation collected by the tactile strength recording unit 14 to determine the relationship between the material physical property value X and the tactile strength. Statistically obtained for each m and defined as a tactile weight vector C (m). The relationship between the obtained tactile movement variable m and the tactile weight vector C (m) is stored in the tactile weight DB 18 and the learning step 1 is completed.
- the Learning step 1 starts by conducting a tactile experiment with a plurality of tentacles 20 as subjects, and obtaining a tactile strength that a person feels from the material.
- the tactile vocabulary specifying unit 9 presents the tactile vocabulary Y in order to instruct the type of tactile sensation that the toucher 20 should touch and feel.
- the tactile vocabulary Y is effectively given as an adjective pair representing opposite tactile sensations such as uneven / flat, smooth / coarse, hard / soft.
- the toucher 20 answers the tactile sensation intensity related to the tactile vocabulary Y felt by touching the learning material 21 as a relative intensity.
- the response method of relative intensity is arbitrary. In the example of FIG. 26, the toucher answers by dividing the relative intensity into five levels. For this reason, all the learning materials 21 are arranged and answered according to the tactile strength when touched. For example, when obtaining tactile sensation strength related to unevenness / flatness, the tactile sensation strength is classified into five levels based on the order and positional relationship of all learning materials 21 rearranged in this order, and the result is answered.
- the tactile sensation strength recording unit 14 records the tactile sensation strength H * of all learning materials for each tactile vocabulary Y as described above.
- the expression of the tactile sensation strength as H * means that the tactile sensation strength is a tactile sensation strength directly answered by the toucher 20.
- the tactile sensation strength H represents the tactile sensation estimated by the material tactile sensation estimation unit 2.
- the tactile sensation strength recording unit 14 records each tactile sensation strength H * in relation to the tactile vocabulary Y presented at this time or the tactile motion feature quantity vector v representing the tactile motion taken by the toucher 20. To do.
- the touch motion of each toucher 20 touching the material is simultaneously measured by the motion physical quantity measuring unit 10, and further converted into the touch motion feature quantity vector v by the motion feature quantity calculating unit 11.
- the details of the movement physical quantity measuring unit 10 and the tactile movement feature quantity vector 11 are exactly the same as in the above-described material search execution. These descriptions are omitted here.
- the tactile movement learning unit 12 collects a large number of tactile movement feature quantity vectors v collected through tactile experiments of all the touchers 20 and divides them into a finite number of characteristic tactile movement groups (clusters). In order to divide a set of many data expressed in vectors as elements into elements that are close to each other and divide them into finite subsets (clusters) based on a predetermined inter-element distance calculation, A technique called clustering is used. The tactile movement learning unit 12 performs clustering using the tactile movement feature amount vector v as an element. The method of dividing the cluster, conditions, and the definition of the distance between elements are arbitrary.
- clustering is performed by setting all collected tactile motion feature vector v as a whole set and using the inter-vector distance of tactile motion feature vector v as an inter-element distance.
- the process is started with the initial state being the number of clusters 1 (whole set) and the total average of the tactile movement feature vector v as the cluster center (H01).
- the number of clusters is n
- one element farthest from the n cluster centers is selected, and this is set as the (n + 1) th cluster center (H02), and all the elements depend on which of the n + 1 cluster centers is the shortest distance from each other.
- Distribute to n + 1 clusters H03
- H04 Distribute to n + 1 clusters
- the inter-vector distance of the tactile motion feature vector v is used as the inter-element distance, but it is also effective to calculate the distance assuming a predetermined distribution such as a normal distribution as the distribution of elements in each cluster. It is. Furthermore, it is also effective to use the similarity of the tactile weight vector C (m) obtained by the tactile weight calculating unit 15 described later as the determination condition for cluster division. By using such values, it is possible to classify tactile movements by considering not only the similarity of movement such as speed and angular velocity but also the similarity of the tactile sensation that is felt as a result. Thus, the tactile sensation strength estimation accuracy can be increased without unnecessarily increasing the tactile movement classification.
- FIG. 28 conceptually shows an example of the result of clustering the tactile motion feature quantity vector v by the method as described above.
- white squares, ⁇ , and ⁇ represent touch motion feature vector v of individual touch motions obtained from toucher 20 by a touch experiment.
- a circle indicated by a broken line indicates a range of each cluster, and each cluster indicates a different tactile movement group.
- Black, ⁇ , ⁇ , and ⁇ indicate the average values of the elements ⁇ , ⁇ , and ⁇ in each cluster, and represent the cluster center used in clustering.
- a symbol such as an index value for identifying the cluster is assigned and stored in the tactile action DB 13 together with information indicating the contents of the cluster.
- a symbol assigned to each cluster is a tactile movement variable m.
- the information indicating the contents of the cluster is information that makes it possible to determine which cluster the arbitrary touch motion feature vector v obtained from the touch motion belongs to. For example, the average vector at the center of the cluster.
- An example of storage in the tactile movement DB 13 is as shown in FIG. 21, for example, as described in the search execution.
- the process proceeds to the construction of the tactile weight DB 18.
- the tactile sensation weight calculation unit 15 the material physical characteristic value measured in advance by the material physical quantity measurement unit 1 for all the learning materials 21 and the tactile sensation strength H * recorded by the tactile strength recording unit 14 through a tactile experiment. Seeking a relationship.
- the measurement means and measurement contents in the material physical quantity measurement unit 1 are the same as those described in the description of the material search unit. Therefore, those descriptions are omitted.
- the tactile sensation weight calculation unit 15 first obtains the relationship between the tactile movement and the tactile vocabulary. That is, the tactile movement feature quantity vector v recorded with the tactile sensation intensity H * is first converted into a tactile movement variable m, and the collected tactile sensation intensity H * and accompanying data are collected for each of the same tactile movement variables. As a result, the most tactile vocabulary Y can be determined for each tactile movement variable m. For example, when the tactile movement variable is m1, and Y1 is the most common tactile vocabulary, the data of the tactile movement variable m1 and the tactile vocabulary Y1 are collected, and the material physical property value X and the tactile intensity H * are In the meantime, the tactile weight vector C (m1) is obtained so that the relationship of Formula 8 is established.
- a tactile weight vector C (m) for estimating (implicitly) the type of tactile sensation felt by the toucher from the tactile movement variable m and calculating the tactile intensity for that tactile type. Can be obtained.
- the tactile weight vector C (m) is sufficient to construct the tactile weight vector C (m), and a large number of learning materials 21, a tactile vocabulary set, and a large number of touchers 20 who respond to the tactile sensation of touching the material (subject test subjects). ) Must be prepared. In order to obtain a more statistically reliable tactile weight vector C (m), prepare as many materials as possible for learning with different physical characteristics, and the tactile vocabulary set also supports various tactile types. It is desirable to prepare as many vocabularies as possible and to collect as many subjects as possible, including differences in age and sex.
- the method for obtaining the tactile weight vector C (m) using the tactile vocabulary data having the largest number of tactile movement variables is the same.
- tactile vocabulary data is not used for learning.
- the learning flow for obtaining the tactile sensation weight vector C (m) is divided into two, and the step for obtaining the relationship between the tactile movement variable m and the tactile vocabulary Y and the step for obtaining the tactile movement variable m and the tactile weight vector C (m) are divided into two. It may be divided into two. That is, a tactile experiment is performed to determine the tactile movement variable m and the most tactile vocabulary Y at that time. Next, the tactile sensation strength H * is obtained by presenting both the tactile vocabulary and the tactile movement to each toucher and asking them to answer the specific tactile vocabulary under the specific tactile movement.
- the material DB 23 is a database of what kind of tactile sensation can be obtained when the material to be searched is touched by various tactile movements, and an example thereof is shown in FIG. 23 as described above. . That is, the material, the tactile movement variable m indicating the type of tactile movement, and the tactile sensation strength corresponding to the tactile movement are associated with each other.
- the flow of the process for building such a database (learning step 2) is as follows.
- the material physical property value X is measured by the material physical quantity measuring unit 1 described above for each search target material 24 to be registered in the database.
- the tactile weight extraction unit 25 extracts a set of tactile weight vectors C (m) corresponding to the tactile motion variable m stored in the tactile weight DB 18 one by one.
- Expression 8 is applied to the extracted tactile weight vector C (m) together with the material physical property value X in the material tactile sensation estimation unit 2 to calculate a tactile intensity H (m) corresponding to the tactile action m.
- the tactile motion variable m extracted by the tactile sensation weight extraction unit 25 and the tactile sensation strength H (m) calculated by the material tactile sensation estimation unit 2 are stored in the material DB 23 as a set.
- the present invention is not limited to this, and the material DB 23 may store the search target material and the material physical property value of the material in association with each other.
- the tactile sensation intensity of each search target material may be calculated using the tactile weight vector determined from the material physical property value and the tactile movement variable. If this method is used, the tactile strength of each search target material must be calculated when executing a search, but the database stores only the material physical property value for each material, so the database size can be reduced. Can be suppressed. (When the tactile sensation intensity is stored in the database, data of [number of search target materials] ⁇ [number of tactile movement variables] is required.)
- the tactile strength data of the material stored in the material DB 23 is not the tactile strength H estimated from the material physical property value X using the tactile weight vector C (m), but the subject toucher answered through the tactile experiment. It may be the tactile strength H * .
- the tactile strength H * it is necessary to conduct a tactile experiment in which the toucher touches all search target materials and answers the tactile sensation strength, but the value obtained by directly quantifying the human sense is used as the tactile strength of each search target material.
- the search results become more convincing.
- the learning material 21 and the search target material 24 can be made the same. Therefore, the tactile weight vector C (m) is also obtained specifically for the set of search target materials 24, and the tactile strength suitable for the search target material can be estimated, thereby improving the accuracy of the material search.
- the material search apparatus 800 simply touches a material having a tactile sensation similar to that of the material to be searched for. You can search. As a result, information retrieval can be performed using a sense of touching an object in electronic commerce or the like. Tactile sensation can be added to the process of confirming products such as clothes and furniture so far, focusing on images, and electronic transactions and information distribution using networks can be more widely spread to general users.
- the type of tactile sensation felt by the toucher can be uniquely estimated from the tactile movement.
- the tactile movements are very similar, and thus the tactile movement variables m expressed discretely may be the same.
- the tactile movement variable m is determined from the tactile movement of the toucher, the type of tactile sensation felt by the toucher cannot be uniquely determined. According to the present embodiment, such a problem can be solved according to the idea of the probability model.
- the right side denominator P (m) is a constant value.
- P (y) in the second term of the molecule is the probability that the type of tactile sensation that the toucher wants to feel at the time of material search is y, and the type of tactile sensation to be felt as a primary approximation is similar. Can be considered as constant values. If the above assumption is made, (Equation 11) can be rewritten into the following (Equation 12).
- A is a constant value under the same tactile movement variable m.
- a material retrieval apparatus based on the principle described above will be described with reference to FIG.
- the same number is assigned to the component having the same function as the corresponding configuration diagram 14 of the first embodiment.
- the difference from the configuration according to the first embodiment (FIG. 14) is that a tactile motion probability calculation unit 26 and a tactile motion probability DB 27 are added, and some components are modified to perform calculations based on this. Is a point.
- an outline of the flow at the time of executing the material search according to FIG. 29 will be described.
- the general flow is the same as that in the first embodiment. That is, the toucher 20 touches the reference material 22, and the tactile strength felt by the toucher is estimated by using the tactile motion at this time and the material physical property value X of the reference material, and the material search is performed based on this. I do.
- the difference between the present embodiment and the first embodiment is that after the tactile movement variable m is determined from the tactile movement of the toucher 20, the tactile weight vector search unit 19 a uses the tactile weight vector C (m , search y) from tactile weight DB 18, that estimates the tactile intensity H y of each tactile type y in material tactile estimating section 2a, and the operation probability retrieval part 26 touch is seeking to learn in advance
- m) of the tactile sensation type y is retrieved from the tactile motion probability DB 27 and a material search is performed using this.
- the tactile movement probability search unit 26 When the tactile movement variable m is determined from the tactile movement of the toucher 20, the tactile movement probability search unit 26 performs the posterior that the type of tactile sensation felt by the toucher is y under the condition that the tactile movement m is observed. The probability P (y
- m) searched by the tactile movement probability search unit 26 may be all P (y
- the tactile motion probability search unit 26 outputs the posterior probabilities P (y
- the touch motion probability DB 27 stores the posterior probability P (y
- the format may be a table lookup format. An example is shown in FIG.
- the tactile weight vector search unit 19a searches the tactile weight DB 18a for a tactile weight vector necessary for estimating the tactile strength for each type of tactile sensation searched by the tactile motion probability search unit 26.
- the tactile sensation type is automatically determined when the tactile movement variable is determined. Therefore, in Equation 8 for estimating the tactile sensation strength, the tactile sensation strength can be estimated from the tactile movement variable m.
- the type of tactile sensation is not determined only by the tactile movement variable m. Since the tactile sensation strength differs for each type of tactile sensation, a tactile weight vector for estimating the tactile sensation must be obtained for each type of tactile sensation. That is, Equation 8 must be rewritten as follows:
- H y means the tactile sensation strength estimated for the tactile sensation type y.
- the tactile weight vector C (m, y) means that the weight matrix is determined depending on the type of tactile sensation and the tactile movement variable m.
- the tactile weight vector search unit 19a applies all tactile sensations y output by the tactile motion probability search unit 26 to the tactile motion variable m obtained by the toucher 20 using such a tactile weight vector C (m, y). Search for and output. At this time, in order to estimate the tactile sensation strength, the tactile sensation weight vector C (m, y) is output in association with the label y indicating the type of tactile sensation.
- the tactile weight DB 18a stores a tactile weight vector C (m, y) determined by the tactile motion variable m and the tactile type.
- the format may be a table lookup format.
- An example of the tactile weight vector DB 18a is shown in FIG. Unlike the tactile weight vector DB 18 illustrated in FIG. 22 of the first embodiment, the tactile weight vector DB 18a changes the tactile weight depending on the type of tactile sensation.
- the material tactile sensation estimation unit 2a uses the material physical property value X of the reference material 22 measured by the material physical quantity measurement unit 1 and the tactile weight vector C (m, y) output from the tactile weight vector search unit 19a to determine the type of tactile sensation. Estimate the tactile strength of each. The estimation is performed based on Equation 16 described above, and each tactile sensation strength Hy is output in association with the label y indicating the type of tactile sensation.
- the material search unit 3a searches the material DB 23a using the tactile intensity group estimated for the reference material 22.
- one tactile sensation strength relating to the type of tactile sensation uniquely determined from the tactile movement variable m is searched based on the difference between the tactile sensation strength of the reference material and the tactile sensation strength of the search target material. That is, the difference in tactile sensation strength for one type of tactile sensation is used as a search score.
- the material DB 23a needs to store the tactile sensation intensity of each search target material for each tactile movement variable m and tactile sensation type y.
- An example of such a material DB 23a is shown in FIG. In the example shown, once the type y of operating variables m and tactile touch each search target material, thereby making it possible to draw the tactile intensity t y of the material.
- the material search unit 3a processes each search target material stored in the material DB 23a (L01). Focusing on the current material (for example, material t), the search score calculation of this material is started (L02). In the search score calculation according to Equation 13, since the score addition is performed based on the difference in tactile sensation for each type of tactile sensation extracted by the tactile movement probability search unit 26, attention is paid to each type of tactile sensation (L03).
- the score to be added is as follows. References and tactile intensity H y of the material obtained in the material tactile estimating section 2a with respect to the tactile type y, distance difference tactile strength t y of operating variables tactile extracted from the material DB 23a m, material related tactile type y t
- m) into a weight value can be used.
- it is effective to use the value of f (x) x, that is, P (y
- the search score S of the material t is determined by adding the scores for all types of tactile sensations extracted by the tactile movement probability estimation unit 26. Since the score S is based on the difference in tactile sensation intensity of the material, the score S is smaller as the material is similar. Therefore, if the search score S is within the predetermined threshold ⁇ (L06), the material t is sufficiently similar to the reference material 22 in tactile sensation, so this material t is output together with the score S as a search result. (L07).
- the search result output method in the search result display unit 7 is the same as in the first embodiment. Therefore, the description thereof is omitted here.
- these learning steps include a learning step 1 for constructing a tactile motion DB 13, a tactile motion probability DB 27, and a tactile weight DB 18a, and a learning step 2 for constructing a material DB 23a.
- description will be given in order.
- FIG. 33 shows a configuration diagram for constructing the tactile motion DB 13, the tactile motion probability DB 27, and the tactile weight DB 18a according to the learning step 1.
- the same elements as those in the corresponding configuration diagram 24 of the first embodiment are given the same symbols, and detailed description thereof is omitted.
- a tactile movement probability learning unit 28 and a tactile movement probability DB 27 are newly added, and the tactile weight learning unit 15a and the tactile weight DB 18a are modified from the tactile weight learning unit 15 and the tactile weight DB 18 in FIG.
- the rough processing flow of these changes is as follows.
- the tactile movement probability learning unit 28 sets the tactile movement variable m and the tactile sensation type y at that time.
- m) of the tactile sensation is calculated and stored in the tactile motion probability DB 27.
- the tactile vocabulary specifying unit 9a also outputs the tactile vocabulary information presented to the toucher to the tactile movement probability learning unit 28, thereby determining the type y of the tactile sensation felt by the toucher and using this And calculate the tactile movement probability.
- the tactile sensation weight calculation unit 15a obtains the relationship between the tactile sensation intensity H * felt by the toucher 20 as a tactile sensation weight vector C (m, y) for the combination of the tactile movement variable m and the tactile sensation type y.
- the relationship between the tactile movement variable m, the tactile sensation type y, and the tactile weight vector C (m, y) is stored in the tactile weight vector DB 18a, and the learning step 1 is completed.
- this embodiment does not uniquely determine the type of tactile sensation felt by the toucher from the tactile movement variable m. Therefore, each value is calculated using the parameter y that determines the type of tactile sensation separately from the tactile movement variable. For this reason, the tactile vocabulary specifying unit 9a not only defines the type of tactile sensation to be felt by presenting the tactile vocabulary to the toucher 20, but also relates to which tactile sensation each response relating to the tactile sensation strength of the toucher. A variable y indicating that is output.
- Tactile vocabulary Y and tactile sensation y are not exactly the same.
- the tactile vocabulary “fluffy” may contain a more basic tactile “softness” as an element. While it is arbitrary how the relationship between the tactile vocabulary presented to the toucher and the type of tactile sensation is established, the simplest method is to determine the most basic set of tactile types that can be considered and This is a method of determining a set of tactile vocabulary corresponding to and using both. Alternatively, perform sensory evaluation experiments in which subjects respond to the strength of various tactile vocabulary in advance, and obtain the basic types of tactile sensation numerically and statistically by performing factor analysis on the results. The relationship between the tactile vocabulary and the tactile sensation may be used.
- the type of tactile sensation felt by the toucher will be shaken, and this may cause the tactile vocabulary and the type of tactile sensation not to be the same.
- it is possible to reduce shaking related to the type of tactile sensation by sufficiently teaching and training the tactile vocabulary to the subject toucher in advance.
- Tactile movement variable data when a large number of touchers 20 touch a plurality of learning materials 21 and feel a plurality of types of tactile sensations is statistically calculated by the tactile movement probability learning unit 28.
- the tactile motion probability learning unit 28 first obtains a conditional probability for the tactile motion variable m from the tactile sensation type y and the number of occurrences of the tactile motion variable m according to the following equation 17.
- N (m, y) represents the number of times that the tactile motion m and the tactile sensation type y occurred simultaneously
- N (y) represents the number of times that the tactile sensation type y occurred.
- y) related to the common tactile movement variable m is collected and applied to Equation 12, thereby calculating the posterior probability P (y
- the result is stored in the tactile movement probability DB 27 in the format illustrated in FIG.
- the tactile weight calculation unit 15a obtains a tactile weight vector C (m, y) shown in Expression 16 indicating the relationship between the material physical property value of the learning material 21 and the tactile intensity answered by the toucher 20.
- the tactile weight vector is determined depending on the tactile movement variable m and the tactile feeling type y. In other words, different tactile sensations of the same material or different tactile movement variables means different tactile sensations.
- the material physical property value X and the tactile intensity H * based on the tactile experiment are obtained for each piece of data in which the condition of the tactile sensation type y and the tactile action variable m are common. It is put together.
- the tactile sensation intensity recording unit 14 records each tactile sensation intensity H * in association with the tactile sensation type y and the tactile movement feature quantity vector v
- the tactile movement feature quantity vector v is referred to the constructed tactile action DB 13. Is converted into a tactile movement variable m.
- the relationship between the material physical property value X and the tactile sensation strength H * shown in Equation 16 is obtained by a technique such as multiple regression analysis. As shown in FIG. 30 as an example, the obtained results are stored in the tactile weight DB 18a in association with the tactile motion variable m, the tactile type y, and the tactile weight vector C (m, y).
- learning step 1 in the present embodiment is completed.
- learning step 2 for constructing the material DB 23a will be described with reference to FIG.
- elements having the same functions as those in FIG. 25 corresponding to the first embodiment are given the same symbols, and description thereof is omitted.
- the difference between the configuration of FIG. 35 according to the present embodiment and the configuration of FIG. 25 according to the first embodiment is that, in the present embodiment, the tactile weight vector stored in the tactile weight DB 18a is the tactile motion variable m and the tactile type. By being determined for each y, the tactile sensation strength of each material is also calculated and stored for each tactile motion variable m and tactile sensation type y.
- the learning step 2 will be described below.
- the tactile weight extraction unit 25a sequentially extracts a combination of the tactile motion variable m as a condition and the tactile type y together with the tactile weight vector C (m, y) at that time from the tactile weight DB 18a.
- the tactile weight vector C (m, y) is applied to the equation 16, and the tactile sensation strength H y (m) under the tactile motion variable m regarding the tactile type y is obtained.
- the obtained tactile sensation strength H y (m) is stored in the material DB 23a in the format shown in FIG. 31 in association with the tactile motion variable m and the tactile sensation type y.
- the material DB 23a may simply store material physical property values for each material, just like the first embodiment, and may calculate the tactile strength each time the material is searched. In this case, an extra calculation time is required when executing the search, but there is an advantage that the storage capacity of the material DB 23a can be reduced.
- touching a reference material for material search is not limited to one time, and material search may be performed through multiple “touches”. In this case, it is effective to narrow down the search result for each “touch” by AND or expand the range by OR.
- the tactile sensation type y contributes to the material search by overlapping it.
- the search result emphasizes the tactile strength related to the type y. This means that if the toucher touches the reference material in order to feel a consistent tactile sensation while changing the tactile movement, the result will be a search result that emphasizes the tactile intensity for that tactile sensation. ing. This matches even when a person touches an object, and enables a natural material search that does not make the toucher feel uncomfortable.
- Embodiment 3 When the principle of the present invention described at the beginning of Embodiment 1 is applied, the tactile sensation is expressed more directly as a vector having tactile sensation intensities of a plurality of tactile sensations as components, thereby changing the tactile sensation due to a change in tactile movement more directly. Can be handled. First, this principle will be described.
- the tactile sensation intensity for each type of tactile sensation felt by the toucher is represented by H k
- the entire tactile sensation can be represented by the following equation as an M-dimensional column vector. This is hereinafter referred to as a “tactile feature vector”.
- the tactile sensation intensity changes according to the tactile movement variable m expressed by Equation 8 can be expressed by the following equation.
- the tactile weight vector so far is represented as a row component in the matrix C (m) of Equation 19, and is a matrix obtained by combining the tactile weight vectors of M types of tactile sensations. ing.
- this C (m) matrix is referred to as a “tactile weight matrix”.
- the meaning of equation 19 is that the tactile sensation of the material felt by the toucher is expressed as one point of the tactile feature vector space, which is obtained by converting the material as one point of the physical feature space with the tactile weight matrix C (m). It can be obtained (see FIG. 36).
- the basis for searching for a material close to the tactile sensation felt by the toucher by performing a material search using the tactile feature vector is based on the following idea.
- the tactile sensation intensity for each type of tactile sensation felt by the toucher varies depending on the tactile movement.
- the tactile person does not judge the tactile sensation only by the tactile intensity of the most prominent tactile sensation, but may also feel the tactile sensation of the relatively weak tactile sensation intensity. That is, the toucher feels the tactile sensation of the material as one point on the tactile sensation space expressed by the tactile feature quantity vector.
- the position of the material in the tactile space changes with the tactile movement, as shown by the fact that the tactile weight matrix of Equation 19 is determined depending on the tactile movement. Therefore, by performing a material search following the change in the tactile sensation feature vector due to the tactile motion, a material search corresponding to the similarity of the tactile sensation felt by the toucher can be performed.
- FIG. 37 shows a state in which the tactile strength related to the surface roughness changes due to the tactile movement with respect to four materials, materials A, B, C, and D, having different surface roughnesses. That is, the tactile sensation intensity felt when the pushing operation is performed changes like a straight line D02 with respect to the change of the material, and the difference in the tactile sensation strength for each material is small and the average tactile sensation strength is also small.
- the tactile sensation strength changes like a straight line D01, and the difference in tactile sensation strength for each material is large, and the average tactile sensation strength is also large. That is, the change accompanying the difference in the tactile movement of the tactile weight matrix C (m) in Equation 8 is expressed as a straight line slope and intercept in this example.
- FIG. 38 exemplifies this, and the distribution of the tactile intensities of the four materials A, B, C, and D is plotted on the horizontal and vertical axes for the two types of tactile types H 1 and H 2. Is a plot of two tactile movements (race movement, push-in movement).
- the distribution of the material in the tactile strength space varies depending on the tactile movement. For example, in the tracing movement, the distribution extends in the horizontal axis direction, and in the pushing movement, the distribution extends in the vertical axis direction. For this reason, when searching for the material having the tactile sensation closest to the material B, when searching based on the tracing operation, the material C is searched based on the distribution by the tracing operation.
- the material D is searched based on the distribution by the pushing operation.
- FIG. 39 The material search apparatus 802 using the tactile feature vector described above will be described with reference to FIGS. 39, 40 and 41.
- FIG. 39 The material search apparatus 802 using the tactile feature vector described above will be described with reference to FIGS. 39, 40 and 41.
- FIG. 39 shows a method for searching for a material.
- elements having the same functions as those in FIG. 14 corresponding to the first embodiment are given the same symbols, and the description thereof is omitted.
- the processing flow at the time of executing material search in the present embodiment is almost the same as that in the first embodiment.
- the difference is that in the present embodiment, as described above, the tactile sensation strength is not estimated only for the specific tactile type determined from the tactile action variable, but for all tactile types to be considered.
- the similarity is calculated using them as vectors.
- the tactile weight matrix search unit 19b searches the tactile weight matrix C (m) shown in Equation 19 from the tactile weight DB 18b. That is, the tactile weight DB 18b stores the tactile movement variable m and the tactile weight matrix C (m) in association with each other.
- the format may be a table lookup format.
- the material tactile sensation estimation unit 2 b applies the tactile sensation weight matrix C (m) and the material physical property value X of the reference material 22 obtained by the material physical quantity measurement unit 1 to obtain the tactile feature vector H. .
- the material DB search unit 3b obtains the distance between the vectors by using the tactile feature vector of each search target material stored in the material DB 23b and the tactile feature vector H obtained by the material tactile estimation unit 2b. Score. As an example, a Euclidean distance between two vectors is possible.
- a score S (T) related to the material T is expressed by the following formula 20 with a distance given a different weight for each vector component (that is, the type of tactile sensation), that is, each tactile strength of the material T stored in the material DB as t i. You may ask for.
- w (i) is a weight function, and it is also effective to use the posterior probability P (y
- the material with the best score (the most similar material) is output as the search result.
- the display method in the search result display unit 7 is the same as that in the first embodiment.
- this learning step includes a learning step 1 for constructing a tactile motion DB 13 and a tactile weight DB 18b, and a learning step 2 for constructing a material DB 23b.
- FIG. 40 is a block diagram for performing learning step 1.
- the learning step 1 of the present embodiment will be described with reference to FIG. Note that elements having the same functions as those in FIG. 24 corresponding to the first embodiment are given the same symbols, and descriptions thereof are omitted.
- a plurality of touchers 20 touch a plurality of learning materials 21 while presenting a plurality of tactile vocabulary words, and a touch experiment is performed to answer the tactile strength.
- all tactile intensities of M types of tactile sensations must be answered for one tactile movement.
- the material tactile sensation estimation unit 2b estimates a vector having the tactile sensation strengths of all M types of tactile sensations as components, and the learning data must also be a vector having the tactile sensation strength as a component. Because.
- the material tactile sensation estimation unit 2 of the first embodiment only estimates the tactile sensation strength of a specific tactile sensation, only the tactile sensation strength of one tactile sensation is answered for one tactile movement during learning. Was good.
- the material retrieval apparatus includes a tactile movement designating unit 29, and the toucher answers the tactile sensation strength for each tactile vocabulary Y after fixing the tactile movement. If a large number of sets of tactile feature vector H * consisting of material physical property value X of learning material, tactile motion variable m, and tactile intensity of M types of tactile sensations are obtained, the relationship of Equation 19 is obtained. These are statistically calculated so that the tactile weight matrix C (m) is obtained by the tactile weight calculating unit 15b. As a result, the tactile motion variable m and the tactile weight matrix C (m) are associated and stored in the tactile weight DB 18b.
- the tactile movement designating unit 29 presents, for example, words for instructing tactile movements such as “tracing” and “pushing in” or a video image of the actual tactile movement.
- the touching motion of the toucher is monitored by the motion physical quantity measuring unit 10 and the motion feature value calculating unit 11, and when a touching motion deviating from a predetermined range is taken, what kind of touching motion is to be taught is taught.
- a warning including a message may be issued.
- an arm having an actuator capable of multi-axis movement including the rotation of the XYZ axis and the rotation of the XYZ axis forcibly holds the hand / finger of the toucher so that a predetermined touch motion is achieved.
- the tactile strength may be answered by moving.
- the tactile movements specified by the tactile movement designating unit 29 in this way are exhaustive for the sake of obtaining the tactile weight matrix C (m).
- the tactile experiment for constructing the tactile movement DB 13 in the learning step 1 and the tactile experiment for constructing the tactile weight DB 18b may be performed individually. That is, first, a tactile experiment is performed by a method similar to the construction method of the tactile motion DB 13 described in the learning step 1 of the first embodiment, and the tactile motion DB 13 is constructed. As a result, a plurality of tactile movement groups are obtained together with the average value of the tactile movement feature quantity vector v of each group.
- the tactile movement designating unit 29 forcibly reproduces the tactile movement of the words, images, or tactile movement feature vector v representing the tactile movement of each tactile movement group by the multi-axis arm. And let the toucher touch the material. In this way, the toucher can take all types of tactile movements comprehensively. By sequentially instructing all combinations of tactile movements and tactile vocabulary in this way, a response regarding the tactile intensity of [number of types of tactile movements] x [number of types of tactile movements] is obtained from each toucher. Can do.
- the tactile weight calculation unit 15b collects tactile intensity data H * relating to each tactile person, each learning material, each tactile motion variable m, and each tactile type y, and statistically processes the collected tactile weight data C *. m) is determined.
- Each row component of the tactile weight matrix C (m) corresponds to one tactile type, which is the same as the tactile weight vector described in the first or second embodiment. That is, for each type of tactile sensation, a tactile weight vector is obtained by the learning method described in the first embodiment or the second embodiment, and synthesized to obtain a tactile weight matrix C (m).
- the tactile weight DB 18b stores the tactile weight matrix C (m) thus obtained in association with the tactile movement variable m.
- the difference between the configuration for the learning step 2 and the configuration of the first embodiment is that what is stored in the tactile weight DB 18b is not a tactile weight vector but a tactile weight matrix, correspondingly.
- the tactile sensation intensity estimated by the material tactile sensation estimation unit 2b is not a tactile sensation intensity (scalar) of a specific tactile sensation type but a tactile sensation intensity vector having a tactile sensation intensity of M types of tactile sensation as a component.
- the process flow is the same as that in the learning step 2 of the first embodiment.
- the tactile weight extraction unit 25b extracts one set of the tactile movement variable m and the tactile weight matrix C (m) one by one from the tactile weight DB 18b.
- the material tactile sensation estimation unit 2b applies the material physical property value X measured by the material physical quantity measurement unit 1 to the search target material 24 to the equation 19 for each tactile motion variable m and tactile weight matrix C (m).
- An estimated tactile feature vector H of this material is obtained.
- the search target material 24, the tactile movement variable m, and the tactile feature quantity vector H estimated at that time are stored in association with each other.
- the tactile intensity of all tactile sensations that the toucher feels at that time is estimated as the tactile sensation feature vector, thereby estimating the tactile sensation that the toucher feels when touching the material It becomes possible.
- by performing material search based on this tactile sensation feature vector it is possible to search for the material closest to the tactile sensation that the toucher feels when taking a certain tactile movement by simply touching the reference material. .
- Embodiments 4 to 7 described below are inventions relating to a tactile display device that presents a tactile sensation as if it were touching a material, and a human tactile action and physical property values of an object corresponding to the tactile action.
- the physical psychological conversion means based on the tactile movement acquired from the database created in advance converts the physical special value into the tactile amount, controls the tactile display actuator based on the tactile amount, Reproduce the tactile sensation when touched.
- FIG. 42 is a block diagram showing the tactile sensation processing apparatus 100 according to the present embodiment.
- the tactile sensation processing apparatus 100 measures and records a tactile sensation feature amount according to the tactile movement of the person 104 who touches the object. Then, the recorded tactile sensation feature value is read out according to the tactile movement of the person 107 who touches the tactile display and reproduced on the tactile display.
- the person 104 touching the test object 105 and the person 107 touching the tactile display unit 103 may be the same person or different persons.
- the person 104 and the person 107 are different persons, there are individual differences in the tactile sensation depending on the sensitivity of the tactile receptor and the experience so far. Further, there are individual differences in the movement of a hand touching an object, and a tactile motion when the person 107 touches the tactile display unit 103 is used to obtain a tactile estimated tactile feature vector F ′ H from a tactile feature recording unit 102 described later.
- an individual difference from the person 104 becomes an error in tactile reproducibility.
- the present invention does not deal with this individual difference, and the difference existing between the person 104 and the person 107 remains as it is, and the person 107 is presented with a tactile sensation.
- the subject 104 and the person 107 are the same person, it is natural that a slight difference occurs between trials. As described above, there is some difference between the state when the subject 104 touches the test object 105 and the state when the person 107 touches the tactile display unit 103, and the present invention is within the range of the difference. To reproduce. Therefore, the subject 104 and the person 107 are arbitrary, and may be the same person or different persons.
- the tactile movement of the person 107 who touches the tactile display represents the type of tactile sensation that he / she (toucher) 107 wants to know among various tactile sensations (such as unevenness, softness, friction, and viscosity) received from the object. ing.
- various tactile sensations such as unevenness, softness, friction, and viscosity
- the tactile sensation processing apparatus 100 includes a tactile sensor unit 101, a tactile feature amount recording unit 102, and a tactile display unit 103, and the tactile sensor unit 101 indicates tactile sensation and tactile movement when a person 104 touches the test object 105.
- the measured tactile sensation is recorded as a tactile sensation feature amount in the tactile sensation feature amount recording unit 102 together with the tactile movement.
- the tactile sensation feature value is read from the tactile sensation feature value recording unit 102 in accordance with the tactile movement of the person 107 who touches the tactile sensation display unit 106 and reproduced by the tactile sensation display unit 103.
- the tactile sensor unit 101 includes a hand movement measuring unit 108, an object physical measuring unit 109, and a physical psychological conversion unit 110.
- the human 104 performs a test using the tactile sensation estimation method considering the tactile movement shown in FIG. A tactile sensation when the object 105 is touched is estimated.
- the hand movement measuring unit 108 measures a touching action of the person 104 touching the test object 105.
- Object physical measuring section 109 measures the physical property vector P H of the test object 105 in accordance with the contact state of a person 104 and the test object 105. Since the contact state between the person 104 and the test object 105 is necessary, the hand movement measurement unit 108 inputs the tactile movement feature quantity vector H to the object physical measurement unit 109.
- the physical psychological conversion unit 110 corresponds to the physical psychological conversion unit 2304 shown in FIG.
- the physical psychological conversion unit 110 converts the physical quantity measured by the object physical measurement unit 109 into a psychological quantity using the function MH calculated in the learning step 2305, and the tactile sensation when the person 104 touches the test target 105. Is estimated.
- the function of the hand movement measurement unit 108 is the same as the function of the motion physical quantity measurement unit 10 of the first embodiment already described. Therefore, the description of the hand movement measurement unit 108 is omitted.
- the object physical measurement unit 109 measures the physical characteristic value of the test object 105. At that time, in order to measure a physical characteristic value related to the touch motion of the person 104, the touch motion feature vector H is acquired from the hand movement measurement unit 108. The measured physical property value is output as Sawado test physical property value vector P H.
- “tactile movement” means “movement of the hand of the person 104 touching the test object 105”
- “tactile test physical characteristic value” is a limited physical characteristic related to the tactile movement. Used in the sense of value.
- the configuration example of the object physical measurement unit 109 is the same as the configuration of the material physical quantity measurement unit 1 of the first embodiment already described. Therefore, the description of the object physical measurement unit 109 is omitted.
- the physical measurement sensor is not particularly limited, and measurement by a non-contact sensor such as a laser displacement meter may be used.
- Physical property values object physical measuring unit 109 measures the position information of the physical sensor 504 (X, Y, Z) together with, and is output as Sawado test physical property value vector P H.
- the physical psychological conversion unit 110 converts the tactile test physical property value vector P H into a tactile motion estimated tactile feature vector F ′ H and estimates the tactile sensation when the person 104 touches the test object 105.
- the conversion formula is generated by the learning step 2305 shown in FIG. 1 and prepared as a function MH .
- the function M H is composed of a plurality of (Equation 8) classified according to the tactile movement feature quantity vector, and is used by switching according to the tactile movement feature quantity vector H that is an output from the hand movement measuring unit 108. Therefore, the calculation function of the physical psychological conversion unit 110 is expressed as follows.
- Breakdown of Sawado test physical property value vector P H is, for example, and the roughness coefficient, elastic modulus, coefficient of static friction, the like dynamic friction coefficient.
- the roughness coefficient expresses the surface unevenness by one coefficient, and the coefficient value increases as the height difference of the surface unevenness increases.
- the elastic modulus is the longitudinal elastic modulus (Young's modulus) that represents deformation against vertical pressure on the object and the transverse elastic modulus (shear) that represents deformation against horizontal pressure (elastic force) on the object. (Elastic modulus).
- the classification of the tactile movement feature vector H is a K-mean method in which the nearest neighbor vector is added to a plurality of category center vectors set as initial values and the category center vector is sequentially updated.
- the tactile movement feature quantity vector H is classified into two clusters as a result of the clustering process.
- Equation 21 is specifically described, an equation such as Equation 22 is formed, for example.
- a is a roughness coefficient
- b is a longitudinal elastic modulus
- c is a transverse elastic modulus
- d is a static friction coefficient
- e is a dynamic friction coefficient.
- FIG. 43 is a diagram illustrating a configuration example of the physical psychology conversion unit 110.
- the table 2401 receives the tactile movement feature quantity vector H and the tactile test physical characteristic value vector P H as inputs, and outputs the tactile movement estimated tactile feature quantity vector F ′ H.
- the tactile motion when the subject touches the n learning objects is classified by clustering processing, and in this example, it is classified into two categories. Further, five physical characteristic value is measured and registered as Sawado test physical property value vector P H.
- the output is a vector expressing the response amount that the subject responded in the sensory evaluation experiment as a factor, and is registered corresponding to each of the n learning objects.
- the tactile receptors have different firing thresholds with respect to the vibration frequency, and therefore, the contact state between the finger and the object determines which tactile receptor is fired.
- the firing of the tactile receptor eventually leads to the occurrence of tactile sensation. Therefore, the factor derived from the sensory evaluation experiment that quantitatively describes this tactile sensation has a strong dependence relationship with the contact state between the finger and the object.
- the tactile sensation changes when the tactile movement changes
- the extracted factors also change when the tactile movement changes. Therefore, if the tactile movement is classified into several patterns and factor analysis is performed for each pattern, there is a high possibility that good factor analysis with a high cumulative contribution rate can be performed.
- the tactile sensation feature amount recording unit 102 records the tactile movement feature amount vector H and the tactile movement estimated tactile feature amount vector F ′ H measured by the tactile sense sensor unit 101.
- the touch motion feature vector H is a touch motion feature vector when the person 104 touches the test object 105.
- the tactile motion estimated tactile feature vector F ′ H is a tactile feature vector that is felt when the person 104 touches the test object 105. Both are measurement data relating to the person 104, and as shown in FIG. 44, the tactile action and the tactile sensation are recorded in pairs, and a correspondence table 2601 of tactile action and tactile sensation is created. In the example of FIG. 44, n sets of the tactile movement feature vector H and the tactile movement estimated tactile feature vector F ′ H are recorded.
- the tactile sensation feature amount recording unit 102 outputs a corresponding tactile sensation tactile feature vector F ′ H according to the tactile movement of the person 107 who touches the tactile sensation display unit 103.
- a person gives vibration stimuli of various frequencies to the tactile receptor to obtain various tactile sensations.
- various tactile sensations felt by the person 104 are temporarily accumulated in the tactile sensation feature amount recording unit 102, and then the corresponding tactile sensation is called according to the tactile movement of the person 107.
- the tactile sensation feature amount recording unit 102 by providing the tactile sensation feature amount recording unit 102, information indicating the tactile sensation feature amount is accumulated. As a result, the tactile sensation corresponding to the feature amount desired by the person 107 can be accurately reproduced in the tactile sensation display unit 103 in proportion to the accumulated amount of tactile feature amount.
- FIG. 45 shows a method of selecting the estimated tactile sensation feature quantity vector F ′ H to be presented to the person 107 from the correspondence table 2601 of the tactile action and the tactile sensation of the subject 104 based on the tactile action feature quantity vector H ′ of the person 107. It is a figure explaining. As will be described in detail later, the tactile movement of the person 107 touching the tactile sensation display unit 103 is measured by the hand movement measuring unit 114 included in the tactile sensation display unit 103 and is output as a tactile movement feature quantity vector H ′.
- the nearest neighbor vector detection unit 2602 compares the n tactile movement feature quantity vectors H and the tactile movement feature quantity vectors H ′ in the tactile action / tactile sense correspondence table 2601 and detects the nearest neighbor vector in the vector space. To do.
- the tactile movement feature quantity vector H104, 2 which is the second data is detected as the nearest neighbor vector. Therefore, the tactile motion estimated tactile feature vector F ′ H 104,2 in the second data is output as the tactile motion estimated tactile feature vector F ′ H.
- the tactile motion estimated tactile feature vector F ′ H may be calculated by a weighted average. That is, as shown in (Equation 23), k tactile motion feature amount vectors H104, x having a small distance from the tactile motion feature amount vector H ′ are detected, and the tactile motion estimated tactile feature amount vector F ′ corresponding thereto is detected. H 104, x is multiplied by the weight wi and added.
- the weight w i is obtained from the reciprocal of the distance between the tactile movement feature vector H ′ and the tactile movement feature vector H104, x, as shown in (Equation 24), and the sum of k of the reciprocal is used with the reciprocal as the numerator. Is calculated for the denominator.
- FIG. 46 shows a tactile motion feature vector space.
- a region 601 shows the distribution of tactile movement feature quantity vectors measured from the person 104
- a region 602 shows the distribution of tactile movement feature quantity vectors measured from the person 107.
- the region 601 person 104
- the motion estimation tactile feature vector F ′ H of the test object 105 can be provided with respect to the movement of the test object 105.
- the test object is detected with respect to the movement of a part of the hand of the person 107. It becomes impossible to accurately provide the tactile motion estimation tactile feature vector F ′ H of 105. Since the nearest neighbor vector detection unit 2602 detects the nearest vector, a tactile motion feature quantity vector 901 outside the area 601 (person 104) is placed at the nearest place in the area 601 as shown in FIG. The tactile movement feature quantity vector 903 is detected in a form that has been mapped and has an error 902.
- the relationship between the tactile motion estimated tactile feature vector F ′ H and the tactile motion feature vector H shown in FIG. 44 may be a function description as shown in (Equation 25).
- Tactile display unit 103 psychophysical converter 111, the actuator control unit 112, the actuator 113 has a movement measuring unit 114 of the hands, Sawado estimated tactile feature quantity vector F 'test object 105 based on H Is reproduced for the person 107.
- the psychophysical conversion unit 111 converts the tactile movement estimated tactile feature vector F ′ H into an actuator control signal D ′ H.
- the conversion formula is the same as that of the psychophysical conversion unit 2104 in FIG. 6 and (Formula 6) is used, but the haptic estimated tactile feature quantity vector F ′ H is selected from a plurality of (Formula 6) classified by tactile movement. Select one based on. This is the same as the physical psychology conversion unit 110 in order to perform factor analysis by classifying it into patterns of tactile movements.
- the tactile movement feature quantity vector of the person 107 who touches the tactile display unit 103 is obtained in advance by the K-mean method or the like. Are classified into a plurality of categories, and factor analysis is performed for each category.
- the relationship between the tactile feature vector F ′ as a factor and the actuator control signal D ′ is represented by a matrix V, and the inverse matrix thereof gives (Equation 6).
- the actuator control unit 112 receives the actuator control signal D ′ H and controls the actuator unit 113.
- the actuator unit 113 may be of any method, but specifically, vibration presentation, force presentation, frictional force presentation, etc. are used as described in the conventional example of FIG. However, in the present invention, since the touch motion of the person 107 is measured, the actuator performs only the necessary motion corresponding to the touch motion, and it is not necessary that all the actuators operate simultaneously. That is, when the finger 1901 performs the tracing operation 1902 as shown in FIG. 7A, the feeling of roughness 1803 and the feeling of friction 1805 are reproduced, and the finger 1901 is pushed in as shown in FIG. 7B. When the operation 1905 is performed, the softness 1804 is reproduced.
- the tactile sensation display 1801 generates a vibration 1807 with the unsteady component 1809 of the ultrasonic transducer to present a roughness 1803, and generates a force distribution 1812 with the steady component 1813 of the ultrasonic transducer.
- a softness and softness 1804 is presented.
- the steady component 1813 also affects the vibration 1807, and the roughness sensation 1803 needs to be corrected.
- the influence on the roughness 1803 is corrected by adjusting the amplitude ratio of the steady component 1813 and the unsteady component 1809.
- the control of the force distribution 1812 since the tactile movement is measured, when only the tracing movement 1902 is performed, the control of the force distribution 1812 is not necessary. Therefore, the steady component 1813 assumes only the generation of the vibration 1807. Thus, the control can be freely performed and the positioning is completely opposite to the conventional example that requires correction.
- the tactile sensation processing apparatus 100 can transmit the tactile sensation felt when the person 104 touches the test object 105 to the person 107, and can feel tactile sensation between remote places without physically carrying the object. I can communicate.
- the person 104 and the person 107 are the same person, tactile archiving is possible, and the tactile sensation of touching the real thing is recorded electronically, and the tactile sensation is reproduced on the tactile display with a time difference. It is possible.
- FIG. 48 is a block diagram showing the configuration of the tactile sensation processing apparatus 1000 according to the present embodiment.
- the physical property value of the object is measured according to the tactile movement of the person touching the tactile display, and the tactile sensation when touching the object is calculated based on the tactile movement and the physical characteristic value of the object.
- a tactile sensation processing apparatus that reproduces a tactile sensation display will be described. Note that portions similar to those of the tactile sensation processing device 100 illustrated in FIG. 42 are denoted by the same reference numerals, and detailed description thereof is omitted.
- the tactile sensation processing apparatus 100 accumulates the tactile feature amount when the person 104 touches the test object.
- the tactile sensation processing apparatus 1000 measures the physical characteristic value of the object according to the tactile movement of the person 107 who touches the tactile display in a situation where the person 104 does not exist, and acquires the tactile feature amount.
- the remote person 107 can acquire the tactile sensation of the test object 105. Become.
- the tactile sensation processing apparatus 1000 includes a tactile sensor unit 1001 and a tactile display unit 103.
- the tactile sensation processing apparatus 1000 is defined as one apparatus including the tactile sensor unit 1001 and the tactile display unit 103, but this is for convenience of description.
- the tactile sensor unit 1001 exists in the vicinity of the test object 105 and the tactile display part 103 exists in the vicinity of the person 107.
- the tactile sensation processing apparatus 1000 is realized as a system constructed so as to include two independent components (the tactile sensor unit 1001 and the tactile display unit 103).
- a physical characteristic value of the test object 105 is measured by a touch action when the person 107 touches the tactile display unit 103, and the person 107 moves to the test object 105 based on the touch action and the physical property value of the target object.
- the tactile sensation when touched is reproduced on the tactile sensation display unit 103.
- the tactile sensor unit 1001 includes an object physical measurement unit 1002 and a physical psychological conversion unit 110, and the person 107 touches the test object 105 from the physical characteristic value of the test object 105 and the touch action of the person 107.
- a tactile motion estimated tactile feature vector F ′ H expressing the tactile feel at the time is calculated.
- the object physical measurement unit 1002 measures the physical characteristic value of the test object 105. At this time, in order to measure a physical characteristic value related to the tactile movement of the person 107, a tactile movement feature quantity vector H ′ is acquired from the hand movement measuring section 114 of the tactile sensation display section 103. The measured physical property value is output as Sawado test physical property value vector P H.
- the physical measurement unit 109 of the tactile sensation processing apparatus 100 shown in FIG. 42 moves the physical measurement sensor 504 using an XYZ stage 501 as shown in FIG. Therefore, when the physical measurement sensor 504 is a single-function sensor that can measure one physical characteristic value, it is replaced and measured in multiple times.
- the physical measurement sensor 504 can measure a plurality of physical property values (specifically, surface shape, elastic property, friction property) with a single measuring instrument as disclosed in Non-Patent Document 8, for example.
- the man-hour for measuring the physical characteristic value can be reduced, and the measurement can be completed at one time.
- the person 107 operates the physical sensor 504 by hand movement when touching the tactile display unit 103.
- the actuator should be driven as soon as possible after the person 107 touches the actuator unit 113. Therefore, it is preferable that the physical measurement of the test object 105, which is a premise for driving the actuator, is completed in a short time. In particular, it is more preferable to end the physical measurement of the test object 105 instantaneously. This is because if the physical measurement of the test object 105 can be completed instantaneously, the tactile sensation display unit 103 can present a tactile sensation in real time according to the movement of the hand when touching the tactile display unit 103. Therefore, it is desirable to use a multifunctional physical measurement sensor that can measure various physical characteristic values of an object without replacing the sensor.
- the tactile sensation processing device 1000 all components are always active after activation, and even if the person 107 is not touching the actuator unit 113, the hand movement measurement unit 114 continues to operate, and the object physical measurement unit
- the tactile motion feature vector H ′ is continuously output to 1002.
- the movement of the physical sensor 504 of the object physical measurement unit 1002 is stopped.
- the hand movement measurement unit 114 captures the movement of the hand, and the physical sensor 504 of the object physical measurement unit 1002 changes its position according to the touch motion feature vector H ′.
- the contact state between 504 and the test object 105 changes according to the movement of the hand of the person 107.
- the tactile test physical characteristic value vector P H measured by the object physical measurement unit 1002 is converted into a tactile motion estimated tactile feature vector F ′ H by the physical psychological conversion unit 110, and this is given to the tactile display unit 103. Finally, the actuator unit 113 is driven, and the person 107 generates a tactile sensation according to the movement of the hand.
- the tactile sensation processing apparatus 1000 measures the physical characteristic value of the test object 105 by the movement of the hand when the person 107 touches the tactile display unit 103, and the person 107 touches the test object 105.
- the tactile sensation is reproduced by the tactile sensation display unit 103. Therefore, it is possible to confirm the tactile sensation felt when the test object 105 is touched in a remote place without physically carrying the object.
- FIG. 49 is a block diagram showing the tactile sensation processing device 1100 according to this embodiment.
- the physical property value of the test object is measured and recorded in advance, the physical property value of the object is read according to the movement of the person's hand touching the tactile display, and the tactile sensation when touching the object is obtained.
- a tactile sensation processing apparatus that reproduces a tactile sensation display will be described.
- the tactile sensation processing device 1100 includes a tactile sensor unit 1101, a tactile feature amount recording unit 102, and a tactile display unit 103.
- the tactile sensation processing unit 1100 simultaneously senses the physical property measurement of the test object 105 performed in advance and the movement of the physical sensor at that time.
- the physical characteristic value of the test object 105 is read by detecting the movement of the physical sensor that is recorded in the feature amount recording unit 102 and similar to the movement of the hand that the person 107 touches the test object 105.
- the tactile sensor unit 1101 includes a physical sensor movement instruction unit 1102, an object physical measurement unit 1103, and a physical psychological conversion unit 110, and the physical sensor moves according to the movement instructed by the physical sensor movement instruction unit 1102, and the test target
- the physical characteristic value of the object 105 is measured, and a tactile motion estimated tactile feature vector F ′ H and a sensor motion feature vector A are output.
- the physical sensor movement instruction unit 1102 instructs the motor control unit 511 of the XYZ stage 2501 to move the physical sensor 2504 as shown in FIG.
- the contents of the instruction cover the movement of the human hand and are preset (programmed) in a memory or the like (not shown) of the physical sensor movement instruction unit 1102.
- the tactile movements of a plurality of people are measured in advance and all of them are used as instruction contents.
- an area 1302 that includes all the measured vector groups 1301 is defined in the tactile movement feature vector space, and then, as shown in FIG. Vector groups that are evenly distributed in 1302 may be set, and these may be designated as physical sensor movements.
- the vector 1303 is one of the uniformly arranged vectors, and is arranged so as to pass through the grid point 1304 and the coordinate origin 1305 set in the region 1302.
- a vector group that is evenly distributed in the region 1302 is set so as to pass through all the lattice points. However, since it becomes complicated if illustrated as it is, only one is illustrated in FIG.
- the measurement is performed within a range in which the test object 105 is not damaged. I do. From the viewpoint of the type of tactile movement of the person 107, the measurement data may be excessive, but the merit that the overlapping of tactile sensation areas as shown in FIG. There is.
- the object physical measurement unit 1103 moves the physical sensor 2504 according to the sensor motion feature vector A from the physical sensor motion instruction unit 1102, measures the physical property value of the test object 105, and performs the tactile test physical property value vector P Output as A. Since the tactile sensation processing device 1100 temporarily stores measurement data in the tactile sensation feature amount recording unit 102, the processing of the tactile sensation sensor unit 1101 and the processing of the tactile sensation display unit 103 are separate. Therefore, the physical sensor 2504 attached to the XYZ stage 2501 may be of a type in which a plurality of single function types are replaced.
- the tactile sensation processing device 1100 comprehensively measures the physical characteristic values of the test target object 105 by moving the physical sensor 2504 by the physical sensor movement instruction unit 1102 and copes with any tactile movement of the person 107. it can. Therefore, in the tactile sensation feature amount space, the region of the feature amount vector that can be output by the tactile sensor unit can include the region of the feature amount vector that can be output by the tactile sensation display unit 103, and tactile sensation reproducibility is improved. Further, no manual operation is required for physical measurement of the test object 105. Thus, the tactile sensation felt when the test object 105 is touched can be confirmed at a remote place without physically carrying the object.
- FIG. 52 is a block diagram showing the tactile sensation processing device 1400 according to the present embodiment.
- a tactile sensation processing apparatus that passively reproduces the tactile sensation when touching an object with the finger touching the tactile sensation without moving is described. Since the fingers do not move, the tactile sensation processing device according to the present embodiment is not provided with the hand movement measuring unit 114 of the tactile sensation processing device 100 of the fourth embodiment. Note that portions similar to those of the tactile sensation processing device 100 illustrated in FIG. 42 are denoted by the same reference numerals, and detailed description thereof is omitted.
- the tactile sensation processing device 1400 includes a tactile sensor unit 101 and a tactile display unit 1401, and a person 107 can feel a tactile sensation when the person 104 touches the test object 105 by touching the tactile display unit 1401. .
- the person 107 does not move the finger touching the tactile display unit 1401, but passively generates a tactile sensation by driving the actuator unit 113.
- a person 107 touching the tactile sensation display unit 103 can freely touch the actuator unit 113, and a person 104 similar to the tactile movement of the person 107 measured by the hand movement measurement unit 114.
- the tactile movement estimated tactile feature vector F ′ H is determined by detecting the tactile movement of the tactile movement.
- the tactile sensation processing device 1400 of the present embodiment uses the tactile motion only in the process of estimating the tactile sensation that the person 104 feels, and there is no hand movement in the process in which the person 107 feels the tactile sensation. Therefore, the tactile sensation processing apparatus 1400 feels only the tactile sensation felt by the person 104 passively.
- Tactile display section 1401, psychophysical converter 111, the actuator control unit 1402 has an actuator unit 113, receives the Sawado estimated tactile feature quantity vector F 'H, to operate the actuator unit 113.
- the actuator control unit 1402 controls the actuator unit 113 so that a finger is brought into contact with the actuator unit 113 in a stationary state and a desired tactile sensation is given to the person 107. That is, the calculation of (Equation 6) is executed, and therefore the matrix V needs to be obtained in advance. This procedure is the same as the case where the input / output characteristics of the actuator unit 2102 in FIG. 6 are modeled by (Equation 5), but in this embodiment, various actuator controls are performed without the person 107 moving his / her finger.
- the signal D ′ is given to the actuator control unit 2103.
- the tactile sensation tactile feature quantity vector F ′ H or the tactile action feature quantity vector H such as the tactile feature quantity recording unit 102. It is possible to interpose data recording means in the middle of a series of flows from the actuator 108 to the actuator unit 113. For example, a tactile motion estimated tactile feature vector F ′ H that is an output of the tactile sensor unit 101 is recorded, and this is recorded. It can be reproduced and reproduced by the tactile display unit 1401.
- the tactile sensation processing device 1400 allows the person 107 to passively sense the tactile sensation felt by the person 104, and can transmit the tactile sensation to a remote place without physically carrying an object.
- the present invention pays attention to the tactile sensation that is sometimes felt in the human tactile sensation, and can provide a material group presenting means for similar tactile sensations based on this and a means for electronically confirming the tactile sensation of the object. .
- the process of confirming the material feeling and texture by sending samples can be reduced in time and cost.
- merchandise returned to consumers has a different texture and texture than expected, but merchandise returns have occurred.
- the tactile sensation can be confirmed on the tactile display, reducing the return rate. Can contribute. By reducing the return rate, it is possible to reduce distribution costs and reduce the environmental burden caused by product transportation.
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- User Interface Of Digital Computer (AREA)
- Position Input By Displaying (AREA)
Abstract
Description
(1) 触感の定量化
(2) 触感センシング
(3) 触感ディスプレイ
しかしこのような素材検索装置には、以下のような問題があった。
本実施形態では、素材検索を行う触者の手指の動き、すなわち触動作を用いて、触者が特に感じる触感の種類とその触感強度を推定して素材検索する装置について述べる。 この装置では、数7を修正した数8によって触感強度Hを推定する。
実施形態1では、触者の感じている触感の種類は、触動作から一意に推定できるとしていた。しかしながら触者の感じている触感の種類が異なっているにも関わらず、触動作が非常に似ているため、離散的に表現された触動作変数mが同一になる場合があると考えられる。このような場合、触者の触動作から触動作変数mが決定されても、触者の感じている触感の種類が一意に決定できない、という問題が起きる。本実施形態によれば、このような問題を、確率モデルの考えに従って解決することが可能になる。
実施形態1の冒頭で説明した本発明の原理を応用すると、複数の触感の種類の触感強度を各成分とするベクトルとして触感を表現することで、より直接的に触動作の変化による触感の変化を扱うことが可能になる。まず、この原理を説明する。
以下に述べる実施形態4から実施形態7は、あたかも素材に触れたかのような触感を呈示する、触感ディスプレイ装置に関する発明で、人の触動作、及び、当該触動作に対応する対象物の物理特性値を計測し、事前に作成されたデータベースから取得した当該触動作に基づく物理心理変換手段で物理特製値を触感量に変換し、当該触感量に基づいて触感ディスプレイのアクチュエータを制御し、対象物に触れた際の触感を再現する。
図48は、本実施形態による触感処理装置1000の構成を示すブロック図である。
図49は、本実施形態による触感処理装置1100を示すブロック図である。
図52は、本実施形態による触感処理装置1400を示すブロック図である。
2 素材触感推定部
3 素材DB検索部
7 検索結果表示部
10 動作物理量計測部
11 動作特徴量算出部
12 触動作学習部
13 触動作DB
14 触感強度記録部
15 触感重み算出部
17 触動作推定部
18 触感重みDB
19 触感重みベクトル検索部
20 触者
22 参照用素材
23 素材DB
24 検索対象素材
25 触感重み抽出部
100 触感処理装置
101 触感センサ部
102 触感特徴量記録部
103 触感ディスプレイ部
104 人
105 テスト対象物
108 手の動き計測部
109 対象物物理計測部
110 物理心理変換部
111 心理物理変換部
112 アクチュエータ制御部
113 アクチュエータ部
114 手の動き計測部
Claims (11)
- 人と対象物との接触状態を計測する動き計測部と、
前記対象物の物理特性値を計測する物理計測部と、
人が任意の対象物に触れたときの接触状態に応じた物理特性値と、前記任意の対象物を触れたときの触感に関する特徴量とに基づき算出される、接触状態毎の前記任意の対象物の物理特性の重みに基づいて、前記物理計測部で計測された前記物理特性値から、触感に関する特徴量を生成する物理心理変換部と
を備えた、触感処理装置。 - 前記物理心理変換部により生成される前記触感に関する特徴量を、前記人が前記対象物に触れた際に感じる触感量として出力する出力部をさらに備えた、請求項1に記載の触感処理装置。
- 前記物理計測部は、前記対象物の複数種類の物理特性に関する物理特性値を計測可能であり、
前記物理計測部は、前記人が前記対象物に触れる接触状態に応じて、計測する物理特性の種類を変化させる、請求項1に記載の触感処理装置。 - 前記動き計測部は、前記人の手が前記対象物と接触している際の手の動きに関する特徴量を計測し、
前記手の動きに関する特徴量は、前記手の位置座標、前記手と前記対象物との接触位置座標、前記位置座標の変化量、前記手の移動速度、前記手の移動加速度のうち、少なくともひとつを含む、請求項1に記載の触感処理装置。 - 前記物理計測部は、前記対象物の物理特性値として、前記対象物の形状および応力のうち少なくともひとつを計測する、請求項1に記載の触感処理装置。
- 前記触感に関する特徴量は、前記被験者の官能応答から抽出した因子として定義される凹凸感、硬軟感、摩擦感、粘性感のうち、少なくともひとつを含む、請求項1に記載の触感処理装置。
- 前記物理心理変換部は、予め作成された、前記対象物の物理特性値、および、前記対象物に触れたときの触感に関する特徴量を対応付けた情報を保持しており、
前記情報は、前記手の動きに関する特徴量、前記対象物の物理特性値、および、前記触感に関する特徴量を対応付けたテーブル、または、前記手の動きに関する特徴量および前記学習対象物の物理特性値を入力として前記触感に関する特徴量が出力される関数である、請求項4に記載の触感処理装置。 - 前記出力部から前記触感に関する特徴量を受け取り、受け取った前記触感に関する特徴量に基づいて力を発生させることにより、その触者に触感を生起させる触感ディスプレイ部をさらに備え、
前記触感ディスプレイ部は、
予め用意された変換規則に基づいて、前記記録部から受け取った前記触感に関する特徴量を制御信号に変換する心理物理変換部と、
前記制御信号に基づいて前記触者の手に力を与える駆動部と、
前記触感ディスプレイ部に触れた前記触者の手の動きに関する特徴量を計測する手の動き計測部と
を備え、前記心理物理変換部は、前記触者の手の動き特徴量に基づいて特定される、前記記録部に記録された前記触感に関する特徴量を受け取る、請求項2に記載の触感処理装置。 - 力を発生させることにより、その触者に触感を生起させる触感ディスプレイ部と、
予め与えられたテスト対象物の物理特性値を計測する物理計測部と、
予め用意された、学習対象物の物理計測値および人が前記学習対象物に触れた際の触感に関する特徴量を対応付けた情報に基づいて、計測された前記テスト対象物の物理特性値を、触感に関する特徴量に変換する物理心理変換部と
を備えた触感処理装置であって、
前記触感ディスプレイ部は、
予め用意された変換規則に基づいて、前記触感に関する特徴量を制御信号に変換する心理物理変換部と、
前記制御信号に基づいて前記触者の手に力を与える駆動部と、
前記触感ディスプレイ部に触れた前記触者の手の動きに関する特徴量を計測する手の動き計測部と
を有し、
前記物理計測部は、計測された前記手の動きに関する特徴量に基づいて、前記テスト対象物の物理特性値を計測する、触感処理装置。 - 物理センサを利用してテスト対象物の物理特性値を計測する物理計測部と、
前記テスト対象物の物理特性値を計測する際において、前記物理センサの動きを指示する物理センサ動き指示部と、
学習対象物の物理特性値、被験者が前記学習対象物に触れたときの手の動きに関する特徴量、および前記対象物を触れた時の触感に関する特徴量を、前記手の動きに関する特徴量に応じて参照する前記物理特性値の種類や重みが変更されるようにあらかじめ対応付けた情報に基づいて、計測された前記物理特性値を前記触感に関する特徴量に変換する物理心理変換部と、
特定された前記触感に関する特徴量を、前記人が前記テスト対象物に触れた際に感じる触感量であるとして出力する出力部と
を備えた、触感処理装置。 - 前記出力部から前記触感に関する特徴量を受け取り、受け取った前記触感に関する特徴量に基づいて力を発生させることにより、その触者に触感を生起させる触感ディスプレイ部をさらに備え、
前記触感ディスプレイ部は、
予め用意された変換規則に基づいて、前記記録部から受け取った前記触感に関する特徴量を制御信号に変換する心理物理変換部と、
前記制御信号に基づいて前記触者の手に力を与え、前記触者に触感を生起させるアクチュエータ部と、
前記アクチュエータ部に触れた前記触者の手の動きに関する特徴量を計測する手の動き計測部と
を備え、前記心理物理変換部は、前記触者の手の動き特徴量に基づいて特定される、前記記録部に記録された前記触感に関する特徴量を受け取る、請求項10に記載の触感処理装置。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201080003361.1A CN102227696B (zh) | 2009-05-21 | 2010-05-21 | 触感处理装置 |
US13/062,567 US8570291B2 (en) | 2009-05-21 | 2010-05-21 | Tactile processing device |
JP2010541647A JP4778591B2 (ja) | 2009-05-21 | 2010-05-21 | 触感処理装置 |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2009122948 | 2009-05-21 | ||
JP2009-122948 | 2009-05-21 | ||
JP2009-136212 | 2009-06-05 | ||
JP2009136212 | 2009-06-05 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2010134349A1 true WO2010134349A1 (ja) | 2010-11-25 |
Family
ID=43126040
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2010/003429 WO2010134349A1 (ja) | 2009-05-21 | 2010-05-21 | 触感処理装置 |
Country Status (4)
Country | Link |
---|---|
US (1) | US8570291B2 (ja) |
JP (1) | JP4778591B2 (ja) |
CN (1) | CN102227696B (ja) |
WO (1) | WO2010134349A1 (ja) |
Cited By (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2011144804A1 (en) * | 2010-05-20 | 2011-11-24 | Nokia Corporation | An apparatus for a user interface and associated methods |
EP2702468A1 (en) * | 2011-04-22 | 2014-03-05 | Immersion Corporation | Electro-vibrotactile display |
JP2014112357A (ja) * | 2012-10-31 | 2014-06-19 | Immersion Corp | 触覚効果を用いてユーザインタフェース上で表面フィーチャをシミュレートする方法及び装置 |
WO2014155727A1 (ja) * | 2013-03-29 | 2014-10-02 | 株式会社牧野フライス製作所 | ワークの加工面評価方法、制御装置および工作機械 |
JP2014216024A (ja) * | 2013-04-26 | 2014-11-17 | イマージョンコーポレーションImmersion Corporation | 触覚使用可能な可変表面のためのシステム及び方法 |
WO2015121971A1 (ja) * | 2014-02-14 | 2015-08-20 | 富士通株式会社 | 触感提供装置、及び、システム |
JP5780368B1 (ja) * | 2013-09-26 | 2015-09-16 | 富士通株式会社 | 駆動制御装置、電子機器、及び駆動制御方法 |
JP2015216456A (ja) * | 2014-05-08 | 2015-12-03 | 富士通株式会社 | 制御信号生成方法、制御信号生成装置及びプログラム |
JP2016122445A (ja) * | 2014-12-19 | 2016-07-07 | イマージョン コーポレーションImmersion Corporation | 触覚フィードバックを伴うオブジェクト操作のためのシステム及び方法 |
JP2016197376A (ja) * | 2015-04-06 | 2016-11-24 | 日本放送協会 | 力覚制御装置及び力覚提示装置 |
JP2017516185A (ja) * | 2014-03-14 | 2017-06-15 | 株式会社ソニー・インタラクティブエンタテインメント | 回転自在に設置されたカメラ付きのゲーミングデバイス |
JP2017199147A (ja) * | 2016-04-26 | 2017-11-02 | 日本電信電話株式会社 | 推定装置、推定方法、およびプログラム |
JP2019160035A (ja) * | 2018-03-15 | 2019-09-19 | 東京瓦斯株式会社 | 触感情報システム、触感情報処理装置およびプログラム |
JP2019168865A (ja) * | 2018-03-22 | 2019-10-03 | 東京瓦斯株式会社 | 情報処理装置、プログラム |
JP2019203886A (ja) * | 2018-05-17 | 2019-11-28 | 花王株式会社 | 触感評価方法および触感計測装置 |
WO2020116085A1 (ja) * | 2018-12-05 | 2020-06-11 | ソニー株式会社 | 推定装置、推定方法、及び推定プログラム |
JP2020155038A (ja) * | 2019-03-22 | 2020-09-24 | トヨタ紡織株式会社 | 情報処理装置、および情報処理装置の制御方法 |
JP2020155037A (ja) * | 2019-03-22 | 2020-09-24 | トヨタ紡織株式会社 | 情報処理装置、および情報処理装置の制御方法 |
JP2020155039A (ja) * | 2019-03-22 | 2020-09-24 | トヨタ紡織株式会社 | 情報処理装置、および情報処理装置の制御方法 |
JP2021022005A (ja) * | 2019-07-24 | 2021-02-18 | トヨタ紡織株式会社 | 情報処理装置および制御方法 |
JP2021022006A (ja) * | 2019-07-24 | 2021-02-18 | トヨタ紡織株式会社 | 情報処理装置 |
JP2021022004A (ja) * | 2019-07-24 | 2021-02-18 | トヨタ紡織株式会社 | データ再生装置、データ生成装置、データ構造、データ再生方法、およびデータ生成方法 |
CN112912821A (zh) * | 2018-08-05 | 2021-06-04 | 联触公司 | 根据触觉特性辅助材料或产品选择的方法和系统 |
JP2023025707A (ja) * | 2021-05-19 | 2023-02-22 | アルプスアルパイン株式会社 | 触覚制御装置、プログラム、触覚制御方法、触覚制御システム、サーバ |
JP7544902B1 (ja) | 2023-04-21 | 2024-09-03 | クラシエ株式会社 | 触感フィードバックシステム、触感フィードバック装置、触感フィードバック方法、および、プログラム |
Families Citing this family (69)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB0901858D0 (en) * | 2009-02-05 | 2009-03-11 | Schoegler Benjaman W | Data input device and sound generating device |
US10564721B2 (en) | 2009-03-12 | 2020-02-18 | Immersion Corporation | Systems and methods for using multiple actuators to realize textures |
US9696803B2 (en) | 2009-03-12 | 2017-07-04 | Immersion Corporation | Systems and methods for friction displays and additional haptic effects |
US9746923B2 (en) | 2009-03-12 | 2017-08-29 | Immersion Corporation | Systems and methods for providing features in a friction display wherein a haptic effect is configured to vary the coefficient of friction |
US10007340B2 (en) | 2009-03-12 | 2018-06-26 | Immersion Corporation | Systems and methods for interfaces featuring surface-based haptic effects |
US8487759B2 (en) | 2009-09-30 | 2013-07-16 | Apple Inc. | Self adapting haptic device |
US20120062518A1 (en) * | 2010-09-09 | 2012-03-15 | Light Blue Optics Ltd | Touch Sensing Systems |
US10013058B2 (en) | 2010-09-21 | 2018-07-03 | Apple Inc. | Touch-based user interface with haptic feedback |
US10120446B2 (en) | 2010-11-19 | 2018-11-06 | Apple Inc. | Haptic input device |
US8830188B2 (en) * | 2011-06-21 | 2014-09-09 | Microsoft Corporation | Infrastructural haptics on wall scale interactive displays |
US10007341B2 (en) * | 2011-06-21 | 2018-06-26 | Northwestern University | Touch interface device and method for applying lateral forces on a human appendage |
US9841331B2 (en) * | 2011-09-24 | 2017-12-12 | President And Fellows Of Harvard College | Artificial skin and elastic strain sensor |
US9178509B2 (en) | 2012-09-28 | 2015-11-03 | Apple Inc. | Ultra low travel keyboard |
US9477909B2 (en) * | 2013-01-09 | 2016-10-25 | SynTouch, LLC | Object investigation and classification |
US10559377B2 (en) | 2013-01-09 | 2020-02-11 | Biomed Concepts Inc. | Graphical user interface for identifying diagnostic and therapeutic options for medical conditions using electronic health records |
US9880623B2 (en) * | 2013-01-24 | 2018-01-30 | Immersion Corporation | Friction modulation for three dimensional relief in a haptic device |
US9672553B2 (en) * | 2013-04-15 | 2017-06-06 | Ebay Inc. | Searchable texture index |
US9466305B2 (en) | 2013-05-29 | 2016-10-11 | Qualcomm Incorporated | Performing positional analysis to code spherical harmonic coefficients |
US20140355769A1 (en) | 2013-05-29 | 2014-12-04 | Qualcomm Incorporated | Energy preservation for decomposed representations of a sound field |
WO2015020663A1 (en) | 2013-08-08 | 2015-02-12 | Honessa Development Laboratories Llc | Sculpted waveforms with no or reduced unforced response |
US9779592B1 (en) | 2013-09-26 | 2017-10-03 | Apple Inc. | Geared haptic feedback element |
US9928950B2 (en) | 2013-09-27 | 2018-03-27 | Apple Inc. | Polarized magnetic actuators for haptic response |
WO2015047356A1 (en) | 2013-09-27 | 2015-04-02 | Bodhi Technology Ventures Llc | Band with haptic actuators |
US10126817B2 (en) | 2013-09-29 | 2018-11-13 | Apple Inc. | Devices and methods for creating haptic effects |
US10236760B2 (en) | 2013-09-30 | 2019-03-19 | Apple Inc. | Magnetic actuators for haptic response |
US9317118B2 (en) * | 2013-10-22 | 2016-04-19 | Apple Inc. | Touch surface for simulating materials |
US10276001B2 (en) | 2013-12-10 | 2019-04-30 | Apple Inc. | Band attachment mechanism with haptic response |
JP2015130168A (ja) * | 2013-12-31 | 2015-07-16 | イマージョン コーポレーションImmersion Corporation | 摩擦拡張制御、及び、タッチコントロールパネルのボタンを摩擦拡張制御部へと変換する方法 |
CN103760324B (zh) * | 2014-01-22 | 2015-07-08 | 东华大学 | 一种材料表面质地的模拟触觉评价方法 |
US9501912B1 (en) | 2014-01-27 | 2016-11-22 | Apple Inc. | Haptic feedback device with a rotating mass of variable eccentricity |
KR102191968B1 (ko) * | 2014-01-28 | 2020-12-16 | 삼성전자주식회사 | 형용사를 이용한 진동 생성 방법 및 디바이스 |
US9922656B2 (en) | 2014-01-30 | 2018-03-20 | Qualcomm Incorporated | Transitioning of ambient higher-order ambisonic coefficients |
US9502045B2 (en) | 2014-01-30 | 2016-11-22 | Qualcomm Incorporated | Coding independent frames of ambient higher-order ambisonic coefficients |
JP6319328B2 (ja) * | 2014-02-14 | 2018-05-09 | 富士通株式会社 | 教育用触感提供装置、及び、システム |
DE112014006608B4 (de) | 2014-04-21 | 2024-01-25 | Apple Inc. | Verfahren, Systeme und elektronische Vorrichtungen zum Bestimmen der Kräfteaufteilung für Multi-Touch-Eingabevorrichtungen elektronischer Vorrichtungen |
US9620137B2 (en) | 2014-05-16 | 2017-04-11 | Qualcomm Incorporated | Determining between scalar and vector quantization in higher order ambisonic coefficients |
US10770087B2 (en) | 2014-05-16 | 2020-09-08 | Qualcomm Incorporated | Selecting codebooks for coding vectors decomposed from higher-order ambisonic audio signals |
US9852737B2 (en) | 2014-05-16 | 2017-12-26 | Qualcomm Incorporated | Coding vectors decomposed from higher-order ambisonics audio signals |
DE102015209639A1 (de) | 2014-06-03 | 2015-12-03 | Apple Inc. | Linearer Aktuator |
US9830782B2 (en) | 2014-09-02 | 2017-11-28 | Apple Inc. | Haptic notifications |
US9747910B2 (en) | 2014-09-26 | 2017-08-29 | Qualcomm Incorporated | Switching between predictive and non-predictive quantization techniques in a higher order ambisonics (HOA) framework |
US10353467B2 (en) | 2015-03-06 | 2019-07-16 | Apple Inc. | Calibration of haptic devices |
US10013059B2 (en) * | 2015-04-03 | 2018-07-03 | Disney Enterprises, Inc. | Haptic authoring tool for animated haptic media production |
AU2016100399B4 (en) | 2015-04-17 | 2017-02-02 | Apple Inc. | Contracting and elongating materials for providing input and output for an electronic device |
WO2017044618A1 (en) | 2015-09-08 | 2017-03-16 | Apple Inc. | Linear actuators for use in electronic devices |
JP2017068350A (ja) * | 2015-09-28 | 2017-04-06 | 株式会社東海理化電機製作所 | 操作入力装置 |
US10180731B2 (en) * | 2015-12-28 | 2019-01-15 | Dassault Systemes Americas Corp. | Virtual clay modeling haptic device |
US10039080B2 (en) | 2016-03-04 | 2018-07-31 | Apple Inc. | Situationally-aware alerts |
US10268272B2 (en) | 2016-03-31 | 2019-04-23 | Apple Inc. | Dampening mechanical modes of a haptic actuator using a delay |
WO2018094850A1 (zh) * | 2016-11-26 | 2018-05-31 | 华为技术有限公司 | 用于测试指纹传感器的测试头 |
JP6874842B2 (ja) * | 2017-07-03 | 2021-05-19 | 日本電気株式会社 | 信号処理装置、信号処理方法およびプログラム |
US10622538B2 (en) | 2017-07-18 | 2020-04-14 | Apple Inc. | Techniques for providing a haptic output and sensing a haptic input using a piezoelectric body |
US11709550B2 (en) * | 2018-06-19 | 2023-07-25 | Sony Corporation | Information processing apparatus, method for processing information, and program |
US10599223B1 (en) | 2018-09-28 | 2020-03-24 | Apple Inc. | Button providing force sensing and/or haptic output |
US10691211B2 (en) | 2018-09-28 | 2020-06-23 | Apple Inc. | Button providing force sensing and/or haptic output |
CN109739345A (zh) * | 2018-11-27 | 2019-05-10 | 南京恩诺网络科技有限公司 | 触觉还原方法、装置及系统 |
CN109710059A (zh) * | 2018-11-27 | 2019-05-03 | 南京恩诺网络科技有限公司 | 感官信息还原方法、装置及系统 |
CN109710060B (zh) * | 2018-11-27 | 2022-04-08 | 南京恩诺网络科技有限公司 | 感官信息录制方法及装置、系统 |
CN109620144A (zh) * | 2018-11-27 | 2019-04-16 | 南京恩诺网络科技有限公司 | 触觉信息处理系统 |
CN109447038A (zh) * | 2018-11-27 | 2019-03-08 | 南京恩诺网络科技有限公司 | 触觉信息估算方法及装置、系统 |
KR102124844B1 (ko) * | 2018-12-31 | 2020-06-22 | 재단법인 대구경북과학기술원 | 촉감인식장치, 시스템 및 방법 |
US11478185B2 (en) | 2019-02-12 | 2022-10-25 | Hill-Rom Services, Inc. | Skin dressing having sensor for pressure ulcer prevention |
US11380470B2 (en) | 2019-09-24 | 2022-07-05 | Apple Inc. | Methods to control force in reluctance actuators based on flux related parameters |
EP4054507B1 (en) * | 2019-11-05 | 2023-12-13 | Reflect Innovation Ltd. | Apparatus, system and method for reducing stress |
WO2022054323A1 (ja) * | 2020-09-09 | 2022-03-17 | ソニーグループ株式会社 | 触覚提示装置、触覚提示システム、触覚提示制御方法及びプログラム |
EP4276730A4 (en) * | 2021-01-08 | 2024-06-26 | DIC Corporation | PRODUCT DELIVERY SYSTEM, PRODUCT DELIVERY METHOD, AND PRODUCT DELIVERY PROGRAM |
US11977683B2 (en) | 2021-03-12 | 2024-05-07 | Apple Inc. | Modular systems configured to provide localized haptic feedback using inertial actuators |
US11809631B2 (en) | 2021-09-21 | 2023-11-07 | Apple Inc. | Reluctance haptic engine for an electronic device |
CN115937043B (zh) * | 2023-01-04 | 2023-07-04 | 南京邮电大学 | 一种触觉辅助点云补全的方法 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2001038958A1 (fr) * | 1999-11-24 | 2001-05-31 | Center For Advanced Science And Technology Incubation, Ltd. | Methode et dispositif de stimulation de sensations tactiles par electricite |
JP2006351012A (ja) * | 2005-06-14 | 2006-12-28 | Samsung Electronics Co Ltd | 触覚情報を効率的に提供するための方法及び装置 |
Family Cites Families (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR2411603A2 (fr) * | 1977-12-19 | 1979-07-13 | Zarudiansky Alain | Dispositif et procede d'enregistrement de restitution et de synthese de sensations tactiles |
US5631861A (en) * | 1990-02-02 | 1997-05-20 | Virtual Technologies, Inc. | Force feedback and texture simulating interface device |
JPH06274094A (ja) | 1993-03-19 | 1994-09-30 | Fujitsu Ltd | シミュレーション装置 |
WO1995020787A1 (en) * | 1994-01-27 | 1995-08-03 | Exos, Inc. | Multimode feedback display technology |
JP3225477B2 (ja) | 1994-06-23 | 2001-11-05 | 日本電信電話株式会社 | 触覚刺激表出方法及び装置と触覚刺激ディスプレイ |
US5959613A (en) * | 1995-12-01 | 1999-09-28 | Immersion Corporation | Method and apparatus for shaping force signals for a force feedback device |
JPH0990867A (ja) | 1995-09-27 | 1997-04-04 | Olympus Optical Co Ltd | 触覚呈示装置 |
AU1328597A (en) * | 1995-11-30 | 1997-06-19 | Virtual Technologies, Inc. | Tactile feedback man-machine interface device |
JPH11203019A (ja) | 1998-01-16 | 1999-07-30 | Fuji Xerox Co Ltd | 触覚ディスプレイ |
KR100261883B1 (ko) * | 1998-04-03 | 2000-07-15 | 정명세 | 질감제시방법 및 장치 |
JP3937684B2 (ja) | 2000-04-20 | 2007-06-27 | 富士ゼロックス株式会社 | 触感呈示装置 |
JP2001312633A (ja) | 2000-04-28 | 2001-11-09 | Dainippon Printing Co Ltd | 仮想体感商品注文システム |
US20020191011A1 (en) * | 2001-06-04 | 2002-12-19 | Firooz Rasouli | Virtual remote touch system |
JP3929881B2 (ja) | 2001-12-04 | 2007-06-13 | 独立行政法人科学技術振興機構 | 皮膚感覚提示装置 |
JP2003337645A (ja) | 2002-05-21 | 2003-11-28 | Seiko Epson Corp | 触覚装置 |
CN1521596A (zh) * | 2003-01-29 | 2004-08-18 | 华 王 | 一种光电式多用途输入外设 |
CA2430193A1 (en) * | 2003-05-28 | 2004-11-28 | Mcgill University | Method and apparatus for synthesizing virtual interaction between rigid and deformable bodies |
CA2430317A1 (en) * | 2003-05-29 | 2004-11-29 | Vincent Hayward | Method and apparatus to record and reproduce tactile sensations |
JP4690203B2 (ja) | 2006-01-13 | 2011-06-01 | 日本電信電話株式会社 | 動き認識システムおよび動き認識方法 |
-
2010
- 2010-05-21 JP JP2010541647A patent/JP4778591B2/ja active Active
- 2010-05-21 US US13/062,567 patent/US8570291B2/en active Active
- 2010-05-21 CN CN201080003361.1A patent/CN102227696B/zh active Active
- 2010-05-21 WO PCT/JP2010/003429 patent/WO2010134349A1/ja active Application Filing
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2001038958A1 (fr) * | 1999-11-24 | 2001-05-31 | Center For Advanced Science And Technology Incubation, Ltd. | Methode et dispositif de stimulation de sensations tactiles par electricite |
JP2006351012A (ja) * | 2005-06-14 | 2006-12-28 | Samsung Electronics Co Ltd | 触覚情報を効率的に提供するための方法及び装置 |
Cited By (41)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2011144804A1 (en) * | 2010-05-20 | 2011-11-24 | Nokia Corporation | An apparatus for a user interface and associated methods |
US9367150B2 (en) | 2010-05-20 | 2016-06-14 | Nokia Technologies Oy | Apparatus and associated methods |
EP2702468A1 (en) * | 2011-04-22 | 2014-03-05 | Immersion Corporation | Electro-vibrotactile display |
EP2702468A4 (en) * | 2011-04-22 | 2015-02-25 | Immersion Corp | ELECTRO-VIBRO-TACTILE DISPLAY |
JP2014112357A (ja) * | 2012-10-31 | 2014-06-19 | Immersion Corp | 触覚効果を用いてユーザインタフェース上で表面フィーチャをシミュレートする方法及び装置 |
US10591994B2 (en) | 2012-10-31 | 2020-03-17 | Immersion Corporation | Method and apparatus for simulating surface features on a user interface with haptic effects |
US10139912B2 (en) | 2012-10-31 | 2018-11-27 | Immersion Corporation | Method and apparatus for simulating surface features on a user interface with haptic effects |
WO2014155727A1 (ja) * | 2013-03-29 | 2014-10-02 | 株式会社牧野フライス製作所 | ワークの加工面評価方法、制御装置および工作機械 |
US10018989B2 (en) | 2013-03-29 | 2018-07-10 | Makino Milling Machine Co., Ltd. | Method of evaluating a machined surface of a workpiece, a controlling apparatus and a machine tool |
JP6062533B2 (ja) * | 2013-03-29 | 2017-01-18 | 株式会社牧野フライス製作所 | ワークの加工面評価方法、制御装置および工作機械 |
JP2014216024A (ja) * | 2013-04-26 | 2014-11-17 | イマージョンコーポレーションImmersion Corporation | 触覚使用可能な可変表面のためのシステム及び方法 |
JP5780368B1 (ja) * | 2013-09-26 | 2015-09-16 | 富士通株式会社 | 駆動制御装置、電子機器、及び駆動制御方法 |
US9400571B2 (en) | 2013-09-26 | 2016-07-26 | Fujitsu Limited | Drive controlling apparatus, electronic device and drive controlling method |
WO2015121971A1 (ja) * | 2014-02-14 | 2015-08-20 | 富士通株式会社 | 触感提供装置、及び、システム |
JPWO2015121971A1 (ja) * | 2014-02-14 | 2017-03-30 | 富士通株式会社 | 触感提供装置、及び、システム |
JP2017516185A (ja) * | 2014-03-14 | 2017-06-15 | 株式会社ソニー・インタラクティブエンタテインメント | 回転自在に設置されたカメラ付きのゲーミングデバイス |
JP2015216456A (ja) * | 2014-05-08 | 2015-12-03 | 富士通株式会社 | 制御信号生成方法、制御信号生成装置及びプログラム |
JP2016122445A (ja) * | 2014-12-19 | 2016-07-07 | イマージョン コーポレーションImmersion Corporation | 触覚フィードバックを伴うオブジェクト操作のためのシステム及び方法 |
JP2016197376A (ja) * | 2015-04-06 | 2016-11-24 | 日本放送協会 | 力覚制御装置及び力覚提示装置 |
JP2017199147A (ja) * | 2016-04-26 | 2017-11-02 | 日本電信電話株式会社 | 推定装置、推定方法、およびプログラム |
JP2019160035A (ja) * | 2018-03-15 | 2019-09-19 | 東京瓦斯株式会社 | 触感情報システム、触感情報処理装置およびプログラム |
JP2019168865A (ja) * | 2018-03-22 | 2019-10-03 | 東京瓦斯株式会社 | 情報処理装置、プログラム |
JP7046667B2 (ja) | 2018-03-22 | 2022-04-04 | 東京瓦斯株式会社 | 情報処理装置、プログラム |
JP7365136B2 (ja) | 2018-05-17 | 2023-10-19 | 花王株式会社 | 触感評価方法および触感計測装置 |
JP2019203886A (ja) * | 2018-05-17 | 2019-11-28 | 花王株式会社 | 触感評価方法および触感計測装置 |
CN112912821A (zh) * | 2018-08-05 | 2021-06-04 | 联触公司 | 根据触觉特性辅助材料或产品选择的方法和系统 |
WO2020116085A1 (ja) * | 2018-12-05 | 2020-06-11 | ソニー株式会社 | 推定装置、推定方法、及び推定プログラム |
JP2020155039A (ja) * | 2019-03-22 | 2020-09-24 | トヨタ紡織株式会社 | 情報処理装置、および情報処理装置の制御方法 |
JP7220108B2 (ja) | 2019-03-22 | 2023-02-09 | トヨタ紡織株式会社 | 情報処理装置、および情報処理装置の制御方法 |
JP2020155038A (ja) * | 2019-03-22 | 2020-09-24 | トヨタ紡織株式会社 | 情報処理装置、および情報処理装置の制御方法 |
JP7220107B2 (ja) | 2019-03-22 | 2023-02-09 | トヨタ紡織株式会社 | 情報処理装置、および情報処理装置の制御方法 |
JP2020155037A (ja) * | 2019-03-22 | 2020-09-24 | トヨタ紡織株式会社 | 情報処理装置、および情報処理装置の制御方法 |
JP7138067B2 (ja) | 2019-03-22 | 2022-09-15 | トヨタ紡織株式会社 | 情報処理装置、および情報処理装置の制御方法 |
JP2021022005A (ja) * | 2019-07-24 | 2021-02-18 | トヨタ紡織株式会社 | 情報処理装置および制御方法 |
JP2021022006A (ja) * | 2019-07-24 | 2021-02-18 | トヨタ紡織株式会社 | 情報処理装置 |
JP7268519B2 (ja) | 2019-07-24 | 2023-05-08 | トヨタ紡織株式会社 | データ再生装置、データ生成装置、データ再生方法、およびデータ生成方法 |
JP7337579B2 (ja) | 2019-07-24 | 2023-09-04 | トヨタ紡織株式会社 | 情報処理装置 |
JP2021022004A (ja) * | 2019-07-24 | 2021-02-18 | トヨタ紡織株式会社 | データ再生装置、データ生成装置、データ構造、データ再生方法、およびデータ生成方法 |
JP2023025707A (ja) * | 2021-05-19 | 2023-02-22 | アルプスアルパイン株式会社 | 触覚制御装置、プログラム、触覚制御方法、触覚制御システム、サーバ |
JP7481419B2 (ja) | 2021-05-19 | 2024-05-10 | アルプスアルパイン株式会社 | 触覚制御装置、プログラム、触覚制御方法、触覚制御システム、サーバ |
JP7544902B1 (ja) | 2023-04-21 | 2024-09-03 | クラシエ株式会社 | 触感フィードバックシステム、触感フィードバック装置、触感フィードバック方法、および、プログラム |
Also Published As
Publication number | Publication date |
---|---|
US8570291B2 (en) | 2013-10-29 |
JPWO2010134349A1 (ja) | 2012-11-08 |
JP4778591B2 (ja) | 2011-09-21 |
CN102227696B (zh) | 2014-09-24 |
US20110157088A1 (en) | 2011-06-30 |
CN102227696A (zh) | 2011-10-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP4778591B2 (ja) | 触感処理装置 | |
US11550395B2 (en) | Mid-air haptic textures | |
US8665076B2 (en) | Tactile feedback method and system, computer program and storage medium that stores the computer program | |
Sripati et al. | A continuum mechanical model of mechanoreceptive afferent responses to indented spatial patterns | |
Fishel et al. | Bayesian exploration for intelligent identification of textures | |
US20090278798A1 (en) | Active Fingertip-Mounted Object Digitizer | |
US20160025615A1 (en) | Method and applications for measurement of object tactile properties based on how they likely feel to humans | |
US9690906B2 (en) | Living object investigation and diagnosis using a database of probabilities pertaining to ranges of results | |
Ndengue et al. | Tactile perception and friction-induced vibrations: discrimination of similarly patterned wood-like surfaces | |
Bueno et al. | A simulation from a tactile device to render the touch of textile fabrics: a preliminary study on velvet | |
Janko et al. | On frictional forces between the finger and a textured surface during active touch | |
Takei et al. | Wearable artificial skin layer for the reconstruction of touched geometry by morphological computation | |
Felicetti et al. | Tactile discrimination of real and simulated isotropic textures by friction-induced vibrations | |
Sun et al. | Predicting fingertip forces by imaging coloration changes in the fingernail and surrounding skin | |
Nymoen | Methods and technologies for analysing links between musical sound and body motion | |
Hu et al. | Analysis of fingertip/fabric friction-induced vibration signals toward vibrotactile rendering | |
Fukui et al. | Hand-shape classification with a wrist contour sensor: Analyses of feature types, resemblance between subjects, and data variation with pronation angle | |
Vivar-Estudillo et al. | Tremor Signal Analysis for Parkinson’s Disease Detection Using Leap Motion Device | |
Wiertlewski | Reproduction of tactual textures: transducers, mechanics and signal encoding | |
Takei et al. | Spiral coil beneath fingertip enhances tactile sensation while tracing surface with small undulations | |
Periyaswamy et al. | Tactile Rendering of Textile Materials | |
Aghanavesi et al. | A review of Parkinson’s disease cardinal and dyskinetic motor symptoms assessment methods using sensor systems | |
Salsedo et al. | Architectural design of the haptex system | |
Jing et al. | Different performances of speech and natural gait in identifying anxiety and depression | |
Xu | Deciphering Contact Interactions and Exploration Strategies Underlying Tactile Perception of Material Softness |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201080003361.1 Country of ref document: CN |
|
ENP | Entry into the national phase |
Ref document number: 2010541647 Country of ref document: JP Kind code of ref document: A |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 10777585 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13062567 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 10777585 Country of ref document: EP Kind code of ref document: A1 |