Generating Localized Haptic Feedback over a Spherical Surface

Patrick Coe

1 a

, Grigori Evreinov

1 b

, Mounia Ziat

2 c

and Roope Raisamo

1 d

The Faculty of Information Technology and Communication Sciences, Tampere University, Tampere, Finland

Dept. of Information Design and Corporate Communication, Bentley University, MA, U.S.A.

Keywords:

High Deﬁnition Haptics, Tangible Mental Images, Virtual Tactile Exciter, Interference Maximum, Actuation

Plate.

Abstract:

The ability to control and manipulate haptic imagery (aka imagining haptic sensations in the mind) makes

use of and extends human vision, allowing “seeing by touch”, exploring, and understanding multidimensional

information. In the purpose of exploring potential tools that can support visuo-haptic imagery, we performed

testing on a spherical surface to investigate whether the placement of actuators at key locations and their

activation at different time offsets can be used to generate dynamic movements of peak vibrations at a given

point and across the curved surface. Through our testing of the spherical structure prototypes, we have found

that offset actuations can be used to magnify vibrations at speciﬁc locations on a spherical surface. The

gathered data show that increased amplitude can be created at a given point across the surface by using the

actuation plate instead of multiple actuators afﬁxed to the curved surface. Our plan is to use these results to

induce dynamic haptic images in a vector format across any surfaces in the future.

1 INTRODUCTION

When interacting with graphical objects through a

tactile surface, people combine visual information

with a tangible surface such as physical buttons on

a keyboard, a mouse, or a haptic device. The force

feedback coming from the surface is used to con-

ﬁrm visual input or present general information in-

stead of sound (Fish, 2002). Still, force feedback

parameters can vary within only a limited range of

the magnitude gradient and time (length of tactile

stimuli). Moreover, in most haptic interfaces which

are based on direct ﬁnger touch, force feedback is

referred to as shared forces (those tangential to the

skin). When skin moves laterally over a sensitive sur-

face, the weight generated based on the pressure ap-

plied (65-100g) produces a contact force that leads

to orthogonal skin deformation (normal to the sur-

face). However, human sense of touch is a more so-

phisticated analyzer of processing dynamic arrays of

force vectors (e.g., when distinguishing the concave

and convex components of surfaces). That is obvi-

ous, when haptic textures and objects are simulated

https://orcid.org/0000-0003-3822-1696

https://orcid.org/0000-0001-7132-8378

https://orcid.org/0000-0003-4620-7886

https://orcid.org/0000-0003-3276-7866

with 3D haptic instruments (Culbertson et al., 2018),

but not yet widely applicable to surface haptics on

touchscreens (Kim et al., 2019) when regular haptic

exciters are used. Therefore, in order to display more

complex vector graphic haptic images than primitive

down sampling based reliefs (Loomis and Lederman,

1986; Krufka et al., 2007), dynamically actuated vir-

tual vibration sources of vector force traveling across

the display surface can be used to convey a higher

bandwidth of information to the user (Evreinova et al.,

2014; Evreinova et al., 2012; Loomis, 1981; Loomis

and Lederman, 1986; Kim et al., 2017; Oakley et al.,

2001; Shin and Choi, 2018).

By providing kinesthetic, proprioceptive, and cu-

taneous information, contact surfaces that are actively

explored by ﬁngers deliver a rich haptic experience to

users. The method generally used for tactile simu-

lation of objects and their surfaces has been to con-

trol each “tactile pixel” (or taxel) laid out in a two-

dimensional array (Vechev et al., 2019). Taxels can-

not be used for high-deﬁnition tactile simulation of

objects, instead they have been used for sparse low-

resolution approximation of interactive surfaces and

virtual stages (Culbertson et al., 2018; Loomis and

Lederman, 1986; Krufka et al., 2007). Yet proceed-

ing from visual principles of perception (Soﬁa and

Jones, 2013; Shin and Choi, 2018), to mimic a most

advanced tactile display technology (Xie et al., 2017),

Coe, P., Evreinov, G., Ziat, M. and Raisamo, R.

Generating Localized Haptic Feedback over a Spherical Surface.

DOI: 10.5220/0010189800150024

In Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 2: HUCAPP, pages

15-24

ISBN: 978-989-758-488-6

might not work for haptic visualization because a sur-

face can be characterized by different physical prop-

erties. These need to be perceived, recognized, and

interpreted through haptic imagination as static, dy-

namic, or virtual (cross-section) array of identical el-

ements. A variety of technological approaches have

been explored for surface simulation and control of

properties (mechanical and acoustic) when explor-

ing and interacting directly or indirectly with virtual

surfaces (Evreinova et al., 2012; Evreinova et al.,

2013), simulating shapes (Evreinova et al., 2014;

Evreinova et al., 2013; Follmer et al., 2013), tex-

ture (Shin and Choi, 2018) properties such as stiff-

ness, curvedness (Follmer et al., 2013; Jang et al.,

2016), friction (M

uller-Rakow et al., 2020), and com-

pliance/elasticity (Mansour et al., 2015).

In this work, we considered properties of the seis-

mic wave propagation and interference, to investigate

the possibility of creating feelable precise high deﬁni-

tion tactile points that could travel across a curved sur-

face and lead to the perception of an apparent tactile

motion (Burtt, 1917; Oakley et al., 2001; Park et al.,

2016; Raisamo et al., 2013). Based on this premise,

fewer actuators would be needed to create a high def-

inition vibrotactile display. This could be achieved by

precisely offsetting any number of given actuations.

Our hypothesis is that the resulting point of construc-

tive wave interference would signiﬁcantly amplify the

level of vibration signal above the ambient noise at

the speciﬁc point of contact. If we are able to know

the necessary actuation offsets which are required to

create a point of maximum constructive wave interfer-

ence dynamically at every given point across the sur-

face, then a matrix of values can be stored and used

to stimulate apparent tactile motion by inducing hap-

tic imagination. An example of haptic imagination

is a music teacher asking a pupil to play the piano on

their desk. The student can imagine and feel the music

piece to be played without using the piano keys. We

can enhance this imagination, for example, with vir-

tual movable haptic vibrations that can be felt moving

across the surface and enabled by sequentially trigger-

ing the respective offsets that create a feelable moving

point of maximum interference. This position of vi-

bration interference could be dynamically positioned

in order to display information to a user in a unique

way. Tactile information could move around a user’s

hand, or a user could be instructed to focus on or fol-

low a moving virtual actuator.

Past work in this area of haptic research has

shown a similar approach of inducing points of vir-

tual actuation across a given surface by taking ad-

vantage of wave properties. For example, Enferad

and others (Enferad et al., 2019) worked on gener-

ating a controlled localized point of stimulation us-

ing voltage modulated signals to actuate piezoelectric

patches across an aluminum beam. They successfully

achieved superposition primarily using voltage phase

modulation. Charles Hudin and his team approached

a similar problem by using time-reversal wave focus-

ing (Hudin et al., 2015). A vibrometer calibrated the

time-reversal wave during a focusing stage, followed

by an actuation signal by an array of 32 actuators

bonded to the underside periphery of a glass plate.

This was successful at creating a precise, localized,

point of haptic stimulation. However, the use of a

closed loop control system raises signiﬁcant difﬁcul-

ties as it limits its practical implementation when a

contact point is hidden or suppressed by a ﬁnger in

consumer devices.

Although our current research focuses on local-

ized haptics over the surface of a hemisphere, we be-

lieve that the data and results gathered will be help-

ful for designing haptics over a multitude of differ-

ent shapes. More innovative and interesting device

form factors are continuously revealed. From curved

edges to ﬂexible displays (Huitema, 2012). As in-

terfaces and displays arrive in increasingly complex

and nonstandard form factors, adaptable tactile out-

put will be necessary for the continued advancements

in haptics in consumer devices. Tactile click buttons

that could once be felt began to rapidly phase out in

favor of relying only on the capacitive touchscreens

found on the present devices. We can see continuing

investment in the tech industry from companies such

as Apple (Parisi and Farman, 2018) focused on the

improvement of haptic feedback. It is clear there is

demand for improved tactile feedback. As form fac-

tors evolve and we move away from the traditional

ﬂat display, understanding how advanced haptic sig-

nals can be used to introduce high ﬁdelity haptics will

continue to increase in relevance.

From the aspect of technical novelty, we take a

simpliﬁed approach over existing research to achieve

localized points of actuation. Although the use of a

multitude of actuators is effective, it is not practical as

it introduces complexity and cost. Furthermore, any

system that needs consistent monitoring of a surface

may be difﬁcult to implement outside of a laboratory

setting. We aim to eliminate the aforementioned is-

sues by both reducing the quantity of actuators needed

to produce a localized point of vibration and calculat-

ing the required offsets for a given material. In ad-

dition, the understanding of wave propagation over

a spherical surface opens the techniques that will be

presented in this paper to the multitude of curved and

molded devices continuously being introduced on the

market.

HUCAPP 2021 - 5th International Conference on Human Computer Interaction Theory and Applications

2 CONCEPT & DESIGN OF THE

SPHERICAL HAPTIC DISPLAY

To test the design concept (Figure 1) explained ear-

lier, we have developed a mockup of a spherical hap-

tic display (Figures 2-6). This preliminary design is

used to test the feasibility of the virtual force actua-

tion, as well as possible ways of optimizing the con-

ﬁguration of actuators assembly with respect to these

forces. Optimization can be adjusted by changing the

layout and the number of actuators, the parameters

of the virtual sources of vector force and the conﬁg-

uration of elementary haptic signals. The prototype

will also be used to measure the resulting construc-

tive wave interference propagation across the curved

display surface. Measurements from related research

have demonstrated that it is possible to achieve ac-

curate localization of increased vibration at a desired

point of contact through controlled offset of multiple

signals (Coe et al., 2019b; Coe et al., 2019a).

Figure 1: The variants of haptic actuators assembly afﬁxed

to the actuation plane. 1-5 - Lofelt L5 (1-4) actuators and

Tectonic exciter TEAX25C10-8HS (5). Red arrows indicate

linear motion, while green arrows indicate angular motion.

Black arrows indicate actuator movement. 8 - Represents

a targeted point of increased magnitude or vibration; 6 - an

actuation plate; 7 - a spherical haptic surface.

To generate a virtual vibration source at a point

on the curved contact display surface, we used a spe-

ciﬁc conﬁguration of powerful unidirectional voice

coil actuators (Tectonics and Lofelt). The concept

is deﬁned as the Volumetric Tactile Display (VTD)

and presented in Figure 1. It consists of construc-

Figure 2: Top view of the ﬁrst spherical prototype with cen-

timeter marks across copper tape used for sensor measure-

ment placement.

tive wave interference propagating sequentially to the

speciﬁc points of contact with the skin. The resulting

point of localized vibration is able to properly mediate

haptic signals by integrating spatially and temporally

discrete sensory inputs.

Figure 3: Side view of the ﬁrst spherical prototype with

centimeter marks across copper tape used for sensor mea-

surement placement.

To gather preliminary data, we built two dome

shaped prototypes. The purpose of each prototype

was to investigate multiple methods of localization

that aim to achieve the same goal. One sphere focused

on wave interference, the other was based on princi-

ples of vector forces. Successfully testing each proto-

type individually was done before attempting to com-

bine both principles in the future. The design of both

prototypes was built within a 116mm diameter poly-

carbonate dome. In both domes wires were run from

the inside out to an external motor controller (L298).

Generating Localized Haptic Feedback over a Spherical Surface

The motor controller used an Arduino DUE chosen

for its high speed of 84Mhz, allowing high preci-

sion outputs and the collection of data at intervals of

5.3µm. A copper strip had been used for a precise cal-

ibration of vector forces (when micro-displacements

over touch surface are measured with the MicroSense

sensor) traveling over the Spherical Haptic Surface

(SHS) from the layout of unidirectional actuators.

Figure 4: Top view of the second spherical prototype with

centimeter marks used for sensor measurement placement.

This construction consists of four Lofelt L5 actuators and a

single Tectonic exciter TEAX25C10-8HS at its base.

The ﬁrst dome was used to test the possible wave

interference of seismic signals generated by actuators

afﬁxed directly to a spherical haptic surface. It con-

sisted of four Tectonic actuators (TEAX1402-8) at-

tached from the inside and placed in vertices of the

tetrahedron (Figure 3 and 2). Localization of the vec-

tor force at the desired point of contact over SHS was

to be achieved through the controlled offset of multi-

ple actuation signals. The controlled offset actuation

aimed to shift the point where constructive wave inter-

ference occurred over the surface of the hemisphere.

In the second dome (Figure 5 and 4) we focused

on testing the principles of a shifting magnitude. We

implemented these effects by varying magnitudes of

lateral and vertical movements. The installed more

powerful next-generation Lofelt technology L5 actu-

ators were afﬁxed to the actuation plate along the X

and Y axis, while a powerful Tectonic exciter was

used to actuate vertically in the direction of the Z-

axis. Using this conﬁguration, we planned to rise up

the resulting force of seismic signals initially inter-

fered in orthogonal directions across from an actua-

Figure 5: Side view of the second spherical prototype with

red markers every two centimeters used for measurement

placement.

tion plate by applying different magnitudes of actua-

tion in the vertical and horizontal axis. Nevertheless,

as displayed in Figure 1, unidirectional haptic actua-

tors could be assembled in a different way to generate

both linear (red) and angular (green) momentum of

forces (torques).

During development of the actuation plate, it was

found that the hydrophobic material (such as the

Gorilla-glass, Teﬂon, and silicone) can affect the

sense of convexity vs. concavity at the point of ﬁn-

ger’s contact. Modifying the thickness of a material

such as glass also has an impact on the vibration that

is to be felt (Xu et al., 2019). The result is promis-

ing for future veriﬁcation through a user study. This

opens new ways of simulating volumetric shapes in

virtual and augmented reality. Thus, the combina-

tion of new material properties and actuation technol-

ogy can allow us to induce complex haptic sensations

which are necessary for developing haptic imagina-

tion in both healthy people and those with perceptual

disabilities.

Besides physical parameters, personal exploration

behavioral features will impact on perceiving multi-

ple tactile information gathered when interacting with

SHS during the perception of mental images of the

objects presented. Therefore, user-centered approach

will be used to further clarify the problems and lim-

itations of the proposed interaction techniques (Fig-

ure 6). The spherical surface as shown complements

the shape of the hand. This allows tactile feedback to

propagate across the palm and ﬁngers.

In the design of the second dome, we focused on

the effects of varying magnitudes of lateral and ver-

tical movements. Due to its shape, wave interference

not only existed and occurred traveling over the sur-

face, but the entire object was moved by the vertical

and horizontal movement generated by attached actu-

HUCAPP 2021 - 5th International Conference on Human Computer Interaction Theory and Applications

Figure 6: Top: Hand at rest placed over spherical prototype

demonstrating a comfortable position Middle: Exploratory

behavior demonstrated by touching with the ﬁnger pads

only. Bottom: Four ﬁngers elongated straight forward, as

if trying to feel the ﬂat edge of a surface.

ators. Therefore, it was possible to magnify a point of

maximum vibration by applying different magnitudes

of actuation in the vertical and horizontal axis. These

magnitude points of maximum could also be used

in combination with wave interference maximums to

amplify tactile signals and focus given vibration de-

tected on the surface.

3 METHODS

We explored two methods to determine the optimal

offset in creating a point of peak magnitude vibra-

tion at around ﬁve centimeters from the base of the

hemisphere. This was measured by traveling verti-

cally along the surface of the sphere from the ﬁrst ac-

tuator (A, Figure 2 and 3). The ﬁrst method consisted

of testing a range of offset vibrations on the sphere be-

tween an initial pair of actuators (AB). After ﬁnding

the offset required to reach a point of maximum vi-

bration interference between these two actuators, we

tested an additional actuator, offsetting this third ac-

tuator against the existing offset pulse of the ﬁrst two

actuators (ABC). This process was repeated for the

fourth actuator, offsetting the fourth actuator against

the existing offset pulse of the previous three actua-

tors (ABCD).

Measurements were taken with a MicroSense sen-

sor (Model 5622-LR Probe, with 0.5 mm x 2.5 mm

sensor). As it is a capacitive sensor, a copper strip was

required to be placed over the surface of the sphere for

measurements to take place. The sensor has an accu-

racy of 0-200um with noise of 3.44 um-rms @5kHz

ampliﬁed with Gauging Electronics until 10V and at-

tenuated to a range of 0 to 5V to make it compati-

ble with the Arduino’s analogous input. The sensor

placement was adjusted to follow the curvature of the

sphere.

4 RESULTS

4.1 Constructive Wave Interference

Figure 7 shows that the addition of the third actua-

tor (C) increased displacement signiﬁcantly while the

addition of the fourth actuator (D) introduced a very

small increase. Speciﬁcally, a possible constructive

wave interference happened between the ﬁrst two ac-

tuators A and B, with actuator A, triggered 2ms be-

fore actuator B, creating a displacement of 183µm.

Triggering the third actuator 3ms after actuator B in-

creased this vibration to 257µm. However, trigger-

ing the fourth actuator provided very little increase

to the peak vibration. The fourth actuator (D) ac-

tivated 24ms before actuator A increased the vibra-

tion to 276µm. That said, actuator (D) can still pos-

itively inﬂuence the overall vibration. A deconstruc-

tive pulse provided by actuating actuator (D) 19ms

before actuator A, reduced the vibration to 203µm.

Because we were unsure whether the actuators in

a spherical setup would interact with each other in the

Generating Localized Haptic Feedback over a Spherical Surface

Figure 7: Measured maximum displacement using offsets

for two (AB), three (ABC) and four (ABCD) actuators.

same way as that of a ﬂat actuation plane, we pro-

ceeded with more thorough testing by trying out every

possible combination of offsets between each of the

four actuators for 15ms before and after each other.

The optimal offset found by this process resulted in

a maximum displacement of 276µm which is identi-

cal to the previous result when all four actuators were

activated.

Figure 8: Measured maximum displacement using offsets

when scanning through all four actuators simultaneously.

Figure 8 shows the maximum displacement when

using the full range offset sweep test to determine off-

sets required to reach a maximum displacement. The

found offsets differ, indicating that there are multiple

possible variances of reaching a peak vibration max-

imum. We also found the offset is the result of ac-

tuators (B) and (C) actuated 5ms after actuator (A)

and actuator (D) actuated 9ms after actuator (C). Al-

though we believe the data gathered using the method

of cycling through every possible combination pro-

vides very accurate offsets, the method is hindered by

the length of time required to measure all combina-

tions along with the amount of data that is required to

be collected.

Based on this data, we believe that while some

waves likely travel across the surface, the semi ﬂex-

ible attachment to the base entails that actuators are

likely pulling the entire object. We cannot only con-

sider wave propagation delay; we need to consider the

movement of the entire dome. Besides calculating the

required offset delays, we must test the ideal magni-

tudes and phase of each signal applied to each actua-

tor.

4.2 Combination of Peak Displacement

Magnitudes

Figure 9: Offset required for maximum peak vibrations for

Lofelt L5 actuators with and without Tectonic actuator (cen-

ter coil).

To address the previous issue related to wave propa-

gation, we performed additional testing by combining

different magnitudes of actuation with offset trigger-

ing. For this test we used the second prototype (Figure

4 and 5) with Lofelt L5 actuators for X and Y vibra-

tions, and a central Tectonic actuator for movement

along the Z axis. More speciﬁcally, we actuated the

Lofelt L5 actuators across the X-axis for 10ms along

with the central actuator for 1ms. This conﬁguration

should reduce the magnitude by which the central ac-

tuator is actuated in relation to the Lofelt L5 actuators.

We tested different offsets to ﬁnd the peak vibration

offset (Figure 10). The data displayed in Figure 9

shows a trend until about the third point when the an-

gle of the sphere begins to be more horizontal. The

implication of this trend is that we are not only expe-

riencing forces attributed by wave interference verti-

cally placed actuators, but also vertical displacement

of the entire hemisphere produced by the horizontally

placed actuator. In the future, we would need to mea-

sure different magnitudes for a set offset rather than a

shifting offset. Magnitude can be adjusted by chang-

ing the size of the pulse or adjusting the voltage pro-

vided to a given actuator. Further testing should mea-

HUCAPP 2021 - 5th International Conference on Human Computer Interaction Theory and Applications

sure the range of these adjustments and their effects

on the resulting vibration.

Figure 10: Time offset required for maximum peak vibra-

tions while activating Lofelt L5 actuator (1) for 10ms and

tectonic actuator (5) for 1ms.

5 DISCUSSION AND FUTURE

WORK

The high-ﬁdelity spherical display opens a new form

of interactivity to a wide range of users. The sphere

encompasses a natural form where a user can rest their

hands on for extended periods of time. As we con-

tinue to see increased access to a multitude of inter-

active technologies, we will need to begin to explore

new intuitive methods of feedback and interaction.

Our visual culture has a strong impact on human

intellectual and creative potential as well as the de-

velopment of perceptual and motor abilities (Kant-

ner et al., 1968). Despite the importance of hap-

tics in development of human perception, once spa-

tial visual representations of distance, size, shape,

and motion have been developed, visual information

tends to dominate over haptic perception (Burtt, 1917)

(Klevberg and Anderson, 2002). Blind and visu-

ally impaired users often lack access to much of the

content available in visual form (Jones et al., 2006).

Coincidentally learning is often shown to improve

with the aid visual feedback, often leaving those with

visual impairment struggling in classrooms. Fortu-

nately, it has been shown that with the introduction of

haptic feedback this gap between visual learning can

begin to be bridged.

The use of haptics in education can be expanded to

aid all students compelled and connected with a sub-

ject, building a bridge between for example the sci-

ences and physical reality. A subject that has been

studied with success by David Grow and his study

of educational robotics (Grow et al., 2007) and more

generally by Michael Pantelios (Pantelios et al., 2004)

with input gloves and force-feedback devices. Much

research available (Hamza Lup and Stefan, 2018;

Fern

andez et al., 2016; Christodoulou et al., 2005;

Minogue and Jones, 2006) would suggest that in-

troducing haptics in the educational environment at

all levels can improve learning among students. A

spherical surface as the one we are experimenting

with can provide a durable polycarbonate surface that

can withstand heavy, repeated use. It also provides

a unique surface for students to explore. It would

be possible to be coupled with a spherical projection

(Ferreira et al., 2014; Zuffo et al., 2014; Zhou et al.,

2019) across the surface that would display an inter-

active image or video that can be explored with tactile

feedback.

As we do not limit our idea to any size and expect

to ﬁnd that we will be able to replicate our ﬁndings in

larger and smaller spherical shapes we open the idea

of spherical haptics to a multitude of use cases. We

imagine a haptic hemisphere in place of an analog

control stick on a game controller. Apart from pro-

viding accurate input, feedback can be manipulated

to create a variety of effects. For example, the texture

could change as you go over rough terrain in a game,

or the direction of an enemy could be made apparent

by the localization of the feedback. We can imagine

a larger sphere could be used to control heavy equip-

ment with accuracy over a 3D space, for example a

crane lifting a concrete slab. A spherical display at the

center of a round table could be used as an interactive

map to help a team to collaborate with localized feed-

back providing additional information that could re-

duce visual overload. Localized feedback could alert

a taskforce of underground structures or other areas

of interest.

Manufacturing technologies continuously move

forward. We are moving away from the rigid limita-

tions of consumer electronics design, as the ongoing

trend of miniaturization along with the introduction

of ﬂexible displays and new molded integrated cir-

cuits mean devices can begin to take any imaginable

shape or form. As we know, many user interfaces al-

ready have large areas of signiﬁcant curvature, such as

a mouse, gamepad, or even a vehicle steering wheel.

By introducing vibration that can be localized at any

point across these surfaces we increase the bandwidth

available to the user, with the ability to introduce new,

more natural, interaction cues.

To delve deeper, we are aware that current virtual

reality headsets incorporate believable visual feed-

back but have yet to incorporate high-ﬁdelity haptic

feedback at a widespread consumer level. Incorpo-

rating this increased level of ﬁdelity to current con-

trollers could bring a new level of immersion to cur-

rent technologies (Al-Sada et al., 2018).

Generating Localized Haptic Feedback over a Spherical Surface

There is also potential for such a spherical device

for use in public spaces. The added beneﬁt of pre-

cise tactile feedback could not only make, for exam-

ple, an information kiosk more widely accessible, but

also help users navigate through the system in a noisy

environment such as a mall (Evreinov and Raisamo,

2002).

Interestingly, our current research ﬁnds that the

offsets used to create localized vibration points hap-

pens within milliseconds of each other. This im-

plies that it may be possible to trigger multiple off-

sets rapidly and sequentially to produce multiple focal

points that may be perceived as simultaneous. This

would provide an additional avenue in the creation of

haptic patters, and potentially used for the develop-

ment of haptic imagination.

The eventual goal in this research is to achieve a

perceivable movable actuation that can be mediated

to any location across the spherical surface. What we

would deﬁne as a virtual haptic actuator. We would

still need to investigate this aspect further to achieve

ideal combinations that produce the most efﬁcient

(easily distinguishable) multiple afferent ﬂows, in-

creasing in strength to a speciﬁc point or along edges

across the sphere’s surface. We would also need to

explore how wave interference can be combined with

differing magnitude combinations to increase the pre-

cision and force of a given vibration across the surface

of the sphere, as well as by taking into account per-

ceptual interference of other receptive ﬁelds that can

affect ﬁngertip tactile sensation (Lakshminarayanan

et al., 2015).

Intermediate materials have been known to en-

hance the sense of touch over an object. Cellophane

ﬁlm, for example, has been used by auto body shops

to examine polishing on automobiles (Sano et al.,

2004). Additionally, the Touch Enhancing Pad (Perry

and Wright, 2009), a patented tool composed of lu-

bricant sandwiched between two thin plastic sheets is

useful for detecting tumors in the breast tissue. There-

fore, it would be of interest to test different materials

in an aim to enhance the perceivable localized feed-

back in our haptic spheres.

As we continue this research, we will get a bet-

ter understanding of the use cases this emerging tech-

nique can provide to users.

6 FUTURE APPLICATIONS

Spherical interfaces do exist (SSI, 2020; Benko et al.,

2008; Daniel et al., 2010; Williamson et al., 2015;

Bolton et al., 2011) yet are still not commonly in use.

The proposed platform for high-ﬁdelity haptic feed-

back opens many possibilities of future interaction by

touch. As the shape of the sphere conforms to the

hand in a natural resting position (Jeannerod, 1984;

McRae and McRae, 1977) it can be used for extended

periods of time as a general computing interface aug-

mented with rich tangible information.

When exploring medical imagery, hidden or ob-

scured entities and deeper structures of palpated ob-

ject either biological (tumor) or physical body (defect

inspection) can be enhanced via localized haptic feed-

back. In the same manner, the standard user interface

could be enhanced, for example by allowing a user to

feel and select icons on a desktop that are underneath

a document they are working on without the need to

minimize or change windows.

The spherical surface requires a much lower range

of motion to interact with, which can be beneﬁcial for

anyone with an injury or illness that impairs move-

ment. The high-ﬁdelity feedback can also help those

with poor vision or no vision to navigate an operating

system using detailed haptic imagery.

The spherical interface is not something we see

limited in size. A larger child size spherical interface

could allow children to explore educational material

in a more immersive style. An adult sized spherical

display could act as a kiosk in a mall providing infor-

mation that can often be confusing on a visual only

ﬂat display, such as orientation or direction. A large

sphere could be used as an interface at the center of a

circular meeting table, allowing users to collaborate

with each other. Often meetings are interrupted to

bring up minor details, for example to let a coworker

know that a ﬁle has been sent, or that they need to

step out. This information could be conveyed using

high deﬁnition haptics, eliminating distracting inter-

ruptions.

A haptic spherical interface also allows interest-

ing new methods for creating secure entry into a de-

vice. For example, a passcode based on identify-

ing localized light-pressure patterns can actively be

shifted throughout the surface yet discovered via lo-

calization. Although the input pattern would remain

the same, the continuous shifting of the physical lo-

cation would make it difﬁcult for a third party to ac-

curately capture the passcode. From an outside per-

spective every physical input of the passcode would

appear as if it were unique.

Overall, we see a wide range of possibilities that

can take advantage of the use of a high-ﬁdelity spher-

ical haptic interface. Unlike many interfaces that cur-

rently exist in the consumer space, we imagine the

spherical haptic interface to be highly adaptable and

open to a plethora of design use cases.

HUCAPP 2021 - 5th International Conference on Human Computer Interaction Theory and Applications

7 CONCLUSIONS

Based on the instrumental measurements of construc-

tive interference of the spherical structure prototypes,

we have found that offset actuations can be used

to magnify vibrations at speciﬁc points on a spher-

ical surface. These magniﬁcations can be created

through a combination of two methods: ﬁrst, through

wave interference where we can use the properties of

constructive wave interference to create an ampliﬁed

point on the surface, and second, through the combi-

nation of peak displacement magnitudes, where dif-

fering forces are applied to the X, Y, and Z axis of an

object to increase forces felt at a certain point across

the surface. The current work demonstrates that a lo-

calized vibration effect is reproducible over a curved

surface. Second, using magnitude combinations, we

obtained preliminary data showing that there is an ef-

fect of increased amplitude at a given point across the

surface.

In this research, we have repeatedly observed that

once offsets are found and set, the resulting output

stays remarkably consistent. Sustainability is impor-

tant, as localization offsets should only need to be

gathered once for a given actuator conﬁguration. It

has also been demonstrated that measured losses due

to attenuation for individual actuators are compen-

sated for by the use of the multiple installed actuators.

The level of localization shown has a potential to

improve users’ immersion in XR environments com-

pared to existing global non-localized vibrations that

induce a blurred sense. This distributed haptic reso-

lution can be compared to visual and auditory propa-

gation ﬁeld. The improvement of haptic ﬁdelity in of

itself aims to improve the user experience in a similar

fashion as other sensory modalities.

This work demonstrates that the use of a virtual vi-

bration point can be achieved over three-dimensional

curved structures. This may allow for the use of fewer

actuators in a variety of feedback interfaces when cre-

ating high-ﬁdelity haptics.

ACKNOWLEDGEMENTS

This work was supported by project Multimodal In-

Vehicle Interaction and Intelligent Information Pre-

sentation (MIVI), funded by Business Finland (grant

8004/31/2018).

REFERENCES

Al-Sada, M., Jiang, K., Ranade, S., Piao, X., H

oglund, T.,

and Nakajima, T. (2018). Hapticserpent: A wearable

haptic feedback robot for vr. pages 1–6.

Benko, H., Wilson, A., and Balakrishnan, R. (2008).

Sphere: Multi-touch interactions on a spherical dis-

play. pages 77–86.

Bolton, J., Kim, K., and Vertegaal, R. (2011). Snowglobe:

A spherical ﬁsh-tank vr display. pages 1159–1164.

Burtt, H. E. (1917). Tactual illusions of movement. Journal

of Experimental Psychology, 2(5):371–385.

Christodoulou, S., Garyfallidou, D., Gavala, M., Ioannidis,

G., Papatheodorou, T., and Stathi, E. (2005). Haptic

devices in virtual reality used for education: Design-

ing and educational testing of an innovative system.

Coe, P., Evreinov, G., and Raisamo, R. (2019a). Gel-based

haptic mediator for high-deﬁnition tactile communi-

cation. pages 7–9.

Coe, P., Farooq, A., Evreinov, G., and Raisamo, R. (2019b).

Generating virtual tactile exciter for hd haptics : A

tectonic actuators’ case study. pages 1–4.

Culbertson, H., Schorr, S., and Okamura, A. (2018). Hap-

tics: The present and future of artiﬁcial touch sensa-

tion. Annual Review of Control, Robotics, and Au-

tonomous Systems, 1.

Daniel, S., Wright, C., and Welland, S. (2010). Spherical

display and control device. US Patent 7,755,605.

Enferad, E., giraud audine, C., Fr

eric, G., Amberg, M.,

and Semail, B. (2019). Generating controlled local-

ized stimulations on haptic displays by modal super-

imposition. Journal of Sound and Vibration, 449.

Evreinov, G. and Raisamo, R. (2002). Information kiosks

for all: issues of tactile access. Proc. WWDU 2002.

Evreinova, T., Evreinov, G., and Raisamo, R. (2012). Eval-

uation of effectiveness of the stickgrip device for de-

tecting the topographic heights on digital maps. In-

ternational Journal of Computer Science and Appli-

cations, 9:61–76.

Evreinova, T., Evreinov, G., and Raisamo, R. (2013). Vir-

tual sectioning and haptic exploration of volumetric

shapes in the absence of visual feedback. Advances in

Human-Computer Interaction, 2013.

Evreinova, T., Evreinov, G., and Raisamo, R. (2014). An

exploration of volumetric data in auditory space. Jour-

nal of the Audio Engineering Society, 62:172–187.

Fern

andez, C., Esteban, G., Conde-Gonz

alez, M., and

Garc

ıa-Pe

nalvo, F. (2016). Improving motivation in

a haptic teaching/learning framework. International

Journal of Engineering Education, 32:553–562.

Ferreira, F., Cabral, M., Belloc, O., Miller, G., Kurashima,

C., Lopes, R., Stavness, I., Anacleto, J., Zuffo, M.,

and Fels, S. (2014). Spheree: A 3d perspective-

corrected interactive spherical scalable display.

Fish, D. E. (2002). Force feedback computer input

and output device with coordinated haptic elements.

US6,337,678B1.

Follmer, S., Leithinger, D., Olwal, A., Hogge, A., and Ishii,

H. (2013). inform: Dynamic physical affordances and

Generating Localized Haptic Feedback over a Spherical Surface

constraints through shape and object actuation. pages

417–426.

Grow, D., Verner, L., and Okamura, A. (2007). Educational

haptics. pages 53–58.

Hamza Lup, F. and Stefan, I. (2018). The haptic paradigm

in education: Challenges and case studies.

Hudin, C., Lozada, J., and Hayward, V. (2015). Localized

tactile feedback on a transparent surface through time-

reversal wave focusing. IEEE transactions on haptics,

Huitema, E. (2012). The future of displays is foldable. In-

formation Display, 28:6–10.

Jang, S., Kim, L., Tanner, K., Ishii, H., and Follmer, S.

(2016). Haptic edge display for mobile tactile inter-

action. pages 3706–3716.

Jeannerod, M. (1984). The timing of natural prehension

movements. Journal of Motor Behavior, 16(3):235–

254. PMID: 15151851.

Jones, G., Minogue, J., Oppewal, T., Cook, M., and Broad-

well, B. (2006). Visualizing without vision at the mi-

croscale: Students with visual impairments explore

cells with touch. Journal of Science Education and

Technology, 15:345–351.

Kantner, L. A., Segall, M. H., Campbell, D. T., and Her-

skovits, M. J. (1968). The inﬂuence of culture on vi-

sual perception. Studies in Art Education, 10(1):68.

Kim, S., Park, G., Kim, S.-C., and Jung, J. (2019). Surface

haptics. pages 421–425.

Kim, S.-C., Han, B.-K., and Kwon, D.-S. (2017). Haptic

rendering of 3d geometry on 2d touch surface based

on mechanical rotation. IEEE Transactions on Hap-

tics, PP:1–1.

Klevberg, G. and Anderson, D. (2002). Visual and hap-

tic perception of postural affordances in children and

adults. Human movement science, 21:169–86.

Krufka, S., Barner, K., and Aysal, T. (2007). Visual to tac-

tile conversion of vector graphics. IEEE transactions

on neural systems and rehabilitation engineering : a

publication of the IEEE Engineering in Medicine and

Biology Society, 15:310–21.

Lakshminarayanan, K., Lauer, A., Ramakrishnan, V., Web-

ster, J., and Seo, N. J. (2015). Application of vibration

to wrist and hand skin affects ﬁngertip tactile sensa-

tion. Physiological reports, 3.

Loomis, J. M. (1981). Tactile pattern perception. Percep-

tion, 10(1):5–27.

Loomis, J. M. and Lederman, S. J. (1986). Handbook of

Perception and Human Performance Volume 1: Sen-

sory processes and perceptiong, volume 1. Wiley-

Interscience, New York, NY, 2nd. edition.

Mansour, N., Fath El Bab, A., and Assal, S. (2015). A novel

sma-based micro tactile display device for elasticity

range of human soft tissues: Design and simulation.

McRae, L. T. and McRae, B. J. (1977). Implements us-

able by persons afﬂicted with arthritis. US Patent

4,035,865.

Minogue, J. and Jones, M. (2006). Haptics in education:

Exploring an untapped sensory modality. Review of

Educational Research - REV EDUC RES, 76:317–

348.

uller-Rakow, A., Hemmert, F., Wintergerst, G., and

Jagodzinski, R. (2020). Reﬂective haptics: Resistive

force feedback for musical performances with stylus-

controlled instruments.

Oakley, I., Brewster, S., and Gray, P. (2001). Communicat-

ing with feeling, pages 61–68.

Pantelios, M., Tsiknas, L., Christodoulou, S., and Pap-

atheodorou, T. (2004). Haptics technology in educa-

tional applications, a case study. JDIM, 2:171–178.

Parisi, D. and Farman, J. (2018). Tactile temporalities: The

impossible promise of increasing efﬁciency and elim-

inating delay through haptic media. Convergence:

The International Journal of Research into New Me-

dia Technologies, page 135485651881468.

Park, J., Kim, J., Oh, Y., and Tan, H. (2016). Rendering

moving tactile stroke on the palm using a sparse 2d

array. volume 9774, pages 47–56.

Perry, D. and Wright, H. (2009). Touch enhancing pad.

Patent No. 4,657,021, Filed April 13th., 1989, Issued

Aug. 24th., 1993.

Raisamo, J., Raisamo, R., and Surakka, V. (2013). Com-

parison of saltation, amplitude modulation, and a hy-

brid method of vibrotactile stimulation. Haptics, IEEE

Transactions on, 6:517–521.

Sano, A., Mochiyama, H., Takesue, N., Kikuuwe, R., and

Fujimoto, H. (2004). Touchlens: Touch enhancing

tool. pages 71 – 72.

Shin, S. and Choi, S. (2018). Geometry-based haptic tex-

ture modeling and rendering using photometric stereo.

pages 262–269.

Soﬁa, K. and Jones, L. (2013). Mechanical and psy-

chophysical studies of surface wave propagation dur-

ing vibrotactile stimulation. Haptics, IEEE Transac-

tions on, 6:320–329.

SSI (2020). Screen solutions international: Spherical pro-

jection displays.

Vechev, V., Zarate, J., Lindlbauer, D., Hinchet, R., Shea,

H., and Hilliges, O. (2019). Tactiles: Dual-mode low-

power electromagnetic actuators for rendering contin-

uous contact and spatial haptic patterns in vr. pages

312–320.

Williamson, J., Sund

en, D., and Bradley, J. (2015). Glob-

alfestival: evaluating real world interaction on a spher-

ical display. pages 1251–1261.

Xie, X., Liu, S., Yang, C., Yang, Z., Liu, T., Xu, J., Zhang,

C., and Zhai, X. (2017). A review of smart materials

in tactile actuators for information delivery. C, 3:38.

Xu, H., Peshkin, M., and Colgate, J. (2019). How the me-

chanical properties and thickness of glass affect tpad

performance.

Zhou, Q., Hagemann, G., Fafard, D., Stavness, I., and Fels,

S. (2019). An evaluation of depth and size perception

on a spherical ﬁsh tank virtual reality display. IEEE

Transactions on Visualization and Computer Graph-

ics, PP:1–1.

Zuffo, M., Ferreira, F., Kurashima, C., Cabral, M., Lopes,

R., Anacleto, J., and Fels, S. (2014). Spheree: An

interactive perspective-corrected spherical 3d display.

HUCAPP 2021 - 5th International Conference on Human Computer Interaction Theory and Applications