ACTIVE Quarterly Meeting
May 2000
Institute of Behavioural Sciences,
University of Derby
Outline
• Touch target study
• Point at target study
• Future work
• Useful Field of View & ACTIVE
ACTIVE display could
potentially be implemented
in a variety of locations
Touch Study Aims
1. How accurately drivers can touch a
target in a variety of locations
– Implications for button size/ACTIVE display
layout
2. Is there a difference in accuracy in
different locations?
– Implications for button size/ACTIVE display
layout in different locations
3. Are there any trends in where people
tend to miss the target? e.g. often top
right quadrant
– implications for varying button shapes
over different locations
4. The time taken to touch a target
– Hand off the wheel/attention off the road
5. Total time to complete task
– Hand off the wheel/attention off the road
Potential Displays
S/W
Touch Study Subjects
• 30 subjects
• Within subjects design
• Age
– Range: 18 - 56
– Mean 29
– SD 9.31
• Gender: 33.3%male 66.7% female
• Driven a car: 93.3%
Seating Buck
• Difficult to do in actual car
• Seating buck mock-up of Daewoo
Nubira driving position
• Greater versatility in terms of camera
positions
Crosshair Board
• Areas 1-7
3 4 5
2 6
S/W
1 7
3 x 3 Arrangement of
Crosshairs (Front View)
Touch Study Camera Set-up
Steering
wheel
Camera
Camera
Crosshair
board Driver
Camera Camera
Steering Wheel Camera
Task Aim
“Aim for the centre of the
crosshairs & be as accurate &
as quick as you can”
Experimental Procedure
Overview
• Completed questionnaire &
introduction, consent form
• Practice session
• Explanation of procedure
• Complete actual study
Touch Study Procedure
1. Subject told which area to ‘touch’
2. Yellow flash seen in mirror
3. Touch first cross
4. Repeat 2 & 3 until completed area
5. Investigator informs next area until completed
randomised presentation of all areas x2
Individual Crosshair Rear View
Example 4
pin point quadrants
• 2mm spacing between concentric
circles
Crosshair Analysis
• Each crosshair target placed under a
camera and fed into a computer software
program
Crosshair Analysis
• Each pin point seen on the target was
then selected using a mouse
• mm from centre of cross
Timing Analysis
• Video analysis of:
– Time taken from flash to end position of
point at target
– Time taken from flash until hand returns
to the wheel
Touch Study Results
Accuracy: All data
30
10
mm
-30 -10 -10 10 30
-30
mm
Area 1
30
20
10
0
-30 -10 -10 10 30
-20
-30
Area 2
30
20
10
0
-30 -10 -10 10 30
-20
-30
Area 3
30
20
10
0
-30 -10 -10 10 30
-20
-30
Area 4
30
20
10
0
-30 -10 -10 10 30
-20
-30
Area 5
30
20
10
0
-30 -10 -10 10 30
-20
-30
Area 6
30
20
10
0
-30 -10 -10 10 30
-20
-30
Area 7
15
10
5
0
-30 -10 -5 10 30
-10
-15
Time taken to touch a Single
target
• Range: 3.27 sec
• Min mean time = 0.16sec
– Area 4, cross 14 (just in front of right
finger on s/w)
• Max mean time = 3.43: Area 2, cross
1 (LHS 2nd area up)
Mean time taken from flash
to target touch
1.7
Mean no. of secs
1.6
1.5
1.4
1.3
1.2
A1 A2 A3 A4 A5 A6 A7
Area
Time taken to complete one
‘touch’ action
• Range: 4.73 sec
• Min mean time = 0.37sec
– Area 4, cross 14 (just in front of right
finger on s/w)
• Max mean time = 5.1sec
– Area 6, cross 9 (RHS 2nd area up)
Mean time taken from flash
to hand returning to s/w
Mean time in secs
2.8
2.6
2.4
2.2
2
1.8
1.6
1.4
1.2
A1 A2 A3 A4 A5 A6 A7
Area
Study
• Pointing more appropriate to ACTIVE
• Pointing is different to touching
– No haptic feedback
– People tend to use their proprioception
and then vision to enhance accuracy
when pointing
Point Study Aims
1. How accurately drivers can point at a
target in a variety of locations
– Implications for button size/ACTIVE
display layout compared to touching
target
2. Is there a difference in accuracy in
different locations?
– Implications for button size/ACTIVE
display layout in different locations
3. Are there any trends in where people
tend to miss the target? e.g. often top
right quadrant
– Implications for varying button shapes
over different locations
4. The time taken to point to a target
– Hand off the wheel/attention off the road
– Driver method of movement to operate
ACTIVE - may make gross movement, then
corrective fine movement
5. Total time to complete task
– Hand off the wheel/attention off the road
Experimental Layout
Point Study Methodology
• Completed questionnaire, introduction &
consent form
• Practice session
• Explanation of procedure
• Complete actual study
Point Study Procedure
1. Subject told which area to ‘point’ at
2. Yellow flash seen in mirror
3. Point at first cross
4. Repeat 2 & 3 until completed area
5. Investigator informs next area until completed
randomised presentation of all areas x2
Measuring Pointing with:
1. Pointing Device: LED/lens attached to
underside of index finger
• Rear camera with infra red filter
recorded where light spot located, in
relation to centre of crosses, during
experiment
Pointing Device
LED & lens
30mm
• Markers attached to 3 finger joints
and fingertip on ‘back’ side of hand
and side nearest to body
• Cameras record
– finger’s position in space using a macro
– fingertip distance from board
Point Study Grids
Preliminary Results For
Point Study
• Study completed - 20Ss: analysis in
progress
• Pointing subject to greater error of
accuracy
• Implies larger ‘button’ sizes than current
research on button sizes
• Also consider vehicle
movement/vibration compared to
literature on size
• People stop at different distances
• Pointing tends to be a quicker action
than touching
Future Work
• Evaluation of Dundee GUI Interface
system
• Layout of interface:
– Grouping
– Menus
– Clutter
– Feedback