No Slide Title

Download Report

Transcript No Slide Title

Human Factors and
Haptic Interfaces
Lynette Jones,
Department of Mechanical Engineering,
Massachusetts Institute of Technology.
Human skin
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
Haptics and Vision
Information
capacity (bits/sec)
Temporal
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
Information Channels in
Human Hand
SENSORY
7 classes of Mechanoreceptor
(17,000 – glabrous skin)
2 classes of Thermoreceptor
4 classes of Nocioceptors
3 classes of Proprioceptor
(0-300 in different muscles)
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
Mechanoreceptors
Fiber
group
Modality
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
Absolute threshold (mm)
Tactile Tuning Curves
1
Meissner’s corpuscles: 20-50 Hz
0.1
Merkel cell: 5-15 Hz
0.01
0.001
Pacinian corpuscles: 60-400 Hz
0.0001
0.1
10
Frequency (Hz)
1000
Perception of Vibrotactile Stimuli –
Frequency and Intensity Interact
 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
PSE (db re: standard)
Equal subjective intensity curves
1
0
-1
-2
-3
-4
Standard 25 Hz
-5
20
30
40
50
60
Comparison frequency (Hz)
70
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)
(Johnson & Phillips, 1981)
Temporal variation often improves
acuity and spatially extensive stimuli
improve performance
Temporal acuity and touch
5 ms
20 ms
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
Haptic Displays
Surface texture of object
Contact in the environment
Moving probe across surface - volume
Compliance/stiffness of object
Mass of object
Thermal 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
Ring finger
Women: 10 N
Men: 18 N
(gain) of force between object and operator
Middle finger
Women: 14 N
Men: 26 N
Index finger
Women: 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 ORGAN
Tendon of Achilles
0-20 in mammalian muscles
Activation force: 0.5 – 1.7 mN
Differential threshold (N)
Force Discrimination
10
1
0.1
6%
0.01
0.1
10
Force (N)
1000
Weight Discrimination
Weber fraction
0.16
0.12
0.08
0.04
0
Rapid
motion
Active
lifting
Reflex
lifting
Static
position
Hand
passive
Perceived force (%MVC)
Fatigue and Perceived Force
100
80
60
40
20
0
0
50
100
150
Time (s)
200
250
300
Force control - Index finger flexor
Coefficient of variation (%)
15
Haptic
12
9
6
Visual
and haptic
3
0
2
4
Force (N)
6
Viscosity Discrimination
Differential threshold (N.s/m)
1000
100
10
1
1
10
100
Reference Viscosity (N.s/m)
1000
Viscosity Discrimination
Weber fraction (%)
60
40
20
19%
0
1
100
Viscosity (N.s/m)
10000
Discriminability Parameters
Variable
Weber fraction
Amplitude of vibrotactile
stimulation (20-300 Hz)
Frequency of vibrotactile
stimulation (5-200 Hz)
Pressure on skin
Force
25%
0.1m
11%
0.4 Hz
12%
2 gm/mm2
7%
Stiffness/compliance
17%
Viscosity
19%
Temperature (cold)
Resolution
2%
30 mN
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
Temperature
Hardness
Size/volume
Weight
Individual Variability
Weber fraction
0.4
0.3
0.2
0.1
0
Subjects
Movement Amplitudes
SD Position (mm)
40
35
30
low
25
20
15
high
10
5
0
1
100
Reference Viscosity (N.s/m)
Movement Velocities
SD Velocity (mm/s)
120
100
low
80
60
high
40
20
0
1
100
Reference Viscosity (N.s/m)
Training for the haptically
inept
Feedback
Optimal
motor strategy
Improve discrimination
but not detection
Interface Mechanical Properties
and Human Operator Performance
Delay (ms)
300
250
200
150
0
500
1000
1500
2000
Interface Stiffness (N/m)
2500
Human Operator
Performance
1
Gain
0.8
0.6
0.4
0.2
0
0
500
1000
1500
2000
Interface stiffness (N/m)
2500
Human Operator
Performance
Time to 0.5 pt on Step
Response Function
750
700
650
600
550
500
450
10
1000
Interface Viscosity (N.s/m)
Human Operator
Performance
1
Gain
0.8
0.6
0.4
0.2
0
10
1000
Interface Viscosity (N.s/m)
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

Elevation in skin temperature (deg
C)
Thermal thresholds
3
2.5
2
1.5
1
0.5
0
0
2
4
6
8
Area (cm sq)
10
12
14
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
Visual
Haptic
Perceived Magnitude
20
3
904 gm
SMALL
15
2.6
2.2
10
LARGE
5
1.8
0
1.4
0
200
400
600
800
350 gm
100
Volume (cc)
Weight (g)
(Masin & Crestoni, 1988)
(Ellis & Lederman, 1993)
Saltation effect
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).
P1/P2 locus
150 ms
110 ms
75 ms
< 25 ms
P3 locus
At short and long time
intervals (20-30 and > 200 ms)
P1 will appear near to or
coincident with P2
As time disparity increases
above 50 ms the first stimulus
travels back to its veridical
locus
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