Human Factors and Haptic Interfaces Lynette Jones, Department of Mechanical Engineering, Massachusetts Institute of Technology.

Human Factors and Haptic Interfaces

Lynette Jones,Department of Mechanical Engineering, Massachusetts Institute of Technology.

Surface of average sized adult human: 1.8 m2

(1000 times that of retinae)

Density: 1250 kg/m3

Weight: 5 kg

Total number of axons converging on CNS: 1.1*106 Retina: 106 axons, Cochlea: 6*104 axons

Human skin

Haptics and Vision

Information Temporal capacity (bits/sec) acuity

Fingertip 102 5 ms

Ear 104 0.01 ms

Eye 106-109 25 ms

Time delays differ for hand and eye – hand is quicker than the eye

SENSORY7 classes of Mechanoreceptor (17,000 – glabrous skin)2 classes of Thermoreceptor4 classes of Nocioceptors3 classes of Proprioceptor (0-300 in different muscles)

Information Channels in Human Hand

Sensory Receptor Properties and Haptic Display Design

Resolution and sensitivity of the sensors Temporal processing characteristics

(adaptation, summation) Spatial features of processing Delays in processing information (0.4-120

m/s)

Tactile Receptor Types (7)

Cutaneous and subcutaneous Fiber ModalityMechanoreceptors group

Meissner corpuscle A, Stroking, fluttering

Merkel disk receptor A, Pressure, texture

Pacinian corpuscle A, Vibration

Ruffini ending A, Skin stretch

Hair tylotrich, hair-guard A, Stroking, fluttering

Hair-down A Light stroking

Field A , Skin stretch

Receptors Processing Vibration, Texture

Tactile Tuning Curves

0.0001

0.001

0.01

0.1

1

0.1 10 1000

Frequency (Hz)

Abs

olut

e th

resh

old

(mm

)

Merkel cell: 5-15 Hz

Meissner’s corpuscles: 20-50 Hz

Pacinian corpuscles: 60-400 Hz

Perceived intensity of vibration varies as a function of frequency as well as amplitude

If the amplitude of vibration is changed while frequency is maintained constant, at certain levels a noticeable change in rate occurs

Perception of Vibrotactile Stimuli – Frequency and Intensity Interact

Equal subjective intensity curves

-5

-4

-3

-2

-1

0

1

20 30 40 50 60 70

PSE

(db

re:

sta

ndar

d)

Comparison frequency (Hz)

Standard 25 Hz

Spatial acuity of the skin Skin movement: 0.1-0.2 mm (Gould et al.,

1979)

Spatial frequency of grating: fingertip can distinguish 40-50 µm in spatial period of 0.7-1.0 mm (Morley et al., 1983)

On smooth glass surface, dot of height 1-3 µm and diameter of 550 µm can be detected by the fingertip (Johansson & LaMotte, 1983)

Temporal variation often improves acuity and spatially extensive stimuli improve performance

(Johnson & Phillips, 1981)

Temporal acuity and touch

Successiveness - 2 stimuli (1 ms duration) must be separated by 5.5 ms to be perceived as 2 at a single locus

Temporal order – 2 successive stimuli at separate sites must be separated by 20 ms in order to determine which site first

20 ms5 ms

Haptic Displays

Surface texture of object Contact in the environmentMoving probe across surface - volumeCompliance/stiffness of objectMass of objectThermal properties of object

Human Perception and Haptic Display Design

Scaling (gain) of information between object and operator

Bandwidth required to present information Mechanical properties of interface Magnitude of delays that are tolerable -

symmetric, modality specific

Force-reflecting Interfaces

Scaling (gain) of force between object and operatorRing finger

Women: 10 NMen: 18 N

Middle fingerWomen: 14 N

Men: 26 N

Index fingerWomen: 19 N

Men: 35 N

Activation force <1 N

Mean force: 0.8-0.9 N

Peak force: 1.8-3.3 N

GOLGI TENDON ORGANTendon of Achilles

       0-20 in mammalian musclesActivation force: 0.5 – 1.7 mN

Force Discrimination

0.01

0.1

1

10

0.1 10 1000

Force (N)

Dif

fere

ntia

l thr

esho

ld (

N)

6%

Weight Discrimination

0

0.04

0.08

0.12

0.16

Active lifting

Reflexlifting

Staticposition

Handpassive

Web

er fr

actio

n

Rapid motion

Fatigue and Perceived Force

0

20

40

60

80

100

0 50 100 150 200 250 300

Time (s)

Per

ceiv

ed fo

rce

(%M

VC

)

0

3

6

9

12

15

Force control - Index finger flexor

2 4 6

Force (N)

Coe

ffic

ien

t of

var

iati

on (

%)

Haptic

Visual and haptic

1

10

100

1000

1 10 100 1000

Dif

fere

nti

al t

hre

shol

d (

N.s

/m)

Reference Viscosity (N.s/m)

Viscosity Discrimination

Viscosity Discrimination

0

20

40

60

1 100 10000

Viscosity (N.s/m)

Web

er f

ract

ion

(%)

19%

Discriminability Parameters

Variable Weber fraction ResolutionAmplitude of vibrotactile 25% 0.1m stimulation (20-300 Hz)Frequency of vibrotactile11% 0.4 Hz stimulation (5-200 Hz)Pressure on skin 12% 2 gm/mm2

Force 7% 30 mN

Stiffness/compliance 17%

Viscosity 19%

Temperature (cold) 2% 0.02oC

Discriminability Parameters

Variable Weber fraction Range

Limb position 7% 5-9%

Limb movement 8% 4-19%

Force 7% 5-12%

Stiffness/compliance 17% 7-26%

Viscosity 19% 10-29%

Hand movements and physical properties

Surface texture

Size/volume

Temperature

Hardness

Weight

0

0.1

0.2

0.3

0.4

Subjects

Web

er fr

acti

on

Individual Variability

0

5

10

15

20

25

30

35

40

1 100

SD

Pos

itio

n (

mm

)

Reference Viscosity (N.s/m)

Movement Amplitudes

low

high

0

20

40

60

80

100

120

1 100

SD

Vel

ocit

y (m

m/s

)

Reference Viscosity (N.s/m)

Movement Velocities

low

high

Training for the haptically inept

FeedbackOptimal motor strategyImprove discrimination but not detection

Interface Mechanical Properties and Human Operator Performance

150

200

250

300

0 500 1000 1500 2000 2500

Interface Stiffness (N/m)

Del

ay (

ms)

Human Operator Performance

0

0.2

0.4

0.6

0.8

1

0 500 1000 1500 2000 2500

Interface stiffness (N/m)

Gai

n

Human Operator Performance

450

500

550

600

650

700

750

10 1000

Interface Viscosity (N.s/m)

Tim

e to

0.5

pt o

n St

ep

Res

pons

e Fu

nctio

n

Human Operator Performance

0

0.2

0.4

0.6

0.8

1

10 1000

Interface Viscosity (N.s/m)

Gai

n

Thermal Interfaces and Perception

Thermoreceptor range: 13-45o C Resolution: 0.02-0.05o C (transient) Thermal discriminability: 2% Neutral thermal sensation: 31-36o C Skin temperature correlates well with

perception of cold not warmth Thermoreceptors are NOT thermometers

Thermal thresholds

0

0.5

1

1.5

2

2.5

3

0 2 4 6 8 10 12 14

Area (cm sq)

Ele

vati

on

in

sk

in t

emp

eratu

re (

deg

C)

Thermal Interfaces

Disadvantages Slow response Poor localization (improves with increases

in temperature – 40° C) Prodigious capacity for adaptation and

summation (intensity and areal extent traded)But could be analogous to color in visual

displays.

Illusions – can they be used to trick the haptic

system?

Size-weight illusions – visual and haptic phenomenon

Saltation effect Use auditory cues to enhance haptic

perception

Size-weight Illusion

0

5

10

15

20

0 200 400 600 800

Weight (g)

Perc

eive

d M

agni

tude

LARGE

SMALL

1.4

1.8

2.2

2.6

3

100Volume (cc)

350 gm

904 gm

Visual Haptic

(Masin & Crestoni, 1988) (Ellis & Lederman, 1993)

Saltation effect

At short and long time intervals (20-30 and > 200 ms) P1 will appear near to or coincident with P2

Warning stimulus (P1) 1 s prior to two taps on skin. Present a stimulus at one location (P2 red) and a second at another (P3 blue).

As time disparity increases above 50 ms the first stimulus travels back to its veridical locus

P1/P2 locus P3 locus

150 ms

110 ms

75 ms

< 25 ms

100 mm

Conclusions

• Spatial and temporal processing characteristics of different modalities (pressure, vibration, temperature, force) vary, but can be used to enhance stimulus presentation in a haptic display

• Cross-modal interactions provide an additional tool for representing object properties