Top Banner
Speech vs. typing A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic
14

A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

Jan 20, 2016

Download

Documents

Jeremy Jõnës
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

Speech vs. typingA comparison of two license plate acquiring

methods

Ubiquitous computingDavid Filipovic

Page 2: A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

Police officers -> always on the move

Acquiring license plate numbers while driving

What is better?

The problem

Page 3: A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

Police officers in charge of safety

We can also help to keep them safe

Why?

Page 4: A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

Assumed: typing has more adverse effects on driving performance

Possible explanation: More time spent looking off road

Assumptions – the hypothesis

Page 5: A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

Conduct a study using the driving simulator Have participants drive and perform under

tasks experimental conditions

Collect data

Approach (1)

Page 6: A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

Driving scenario: 2-lane highway Vehicles approached from their right

Primary task: driving Secondary task: collecting license plates

Approach (2)

Page 7: A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

Driving performance data◦ Lane position, steering wheel angle, PDT…

Physiological data◦ EKG, skin conductance, pupil diameter

Qualitative data◦ Participants’ opinions, NASA-TLX

Data collected

Page 8: A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

A ruggedized keyboard for the typing task

Ambient speech recognition (100% accuracy)

The setup

Page 9: A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

Four male subjects

Aged 22-25, average: 23.75

Experienced drivers

Proficient QWERTY keyboard users

Participants

Page 10: A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

The results (1)

Speech Typing0

10

20

30

40

50

60

NASA-TLX Workload rat-ing

Interaction type

NA

SA

-TLX

work

load r

ati

ng

Speech Typing0

1

2

3

4

5

6

7

Task completion time [s]

Interaction type

task c

om

ple

tion t

ime [

s]

Page 11: A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

The results (2)

Speech Typing0

0.010.020.030.040.050.060.070.080.09

Mean lane position variance

Interaction type

mean lane p

osit

ion v

ari

ance

Speech Typing0

0.51

1.52

2.53

3.54

Mean steering wheel angle

variance

Interaction type

mean s

teeri

ng w

heel angle

vari

ance

ANOVA, p<0.022

Page 12: A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

The results (3)

800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000-4

-3

-2

-1

0

1

2

3

4

5

800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000-4

-3

-2

-1

0

1

2

3

4

5

speech typing

Steering wheel angle

Page 13: A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

Definitive presence of an effect

Future study with more participants

Conclusion

Page 14: A comparison of two license plate acquiring methods Ubiquitous computing David Filipovic.

Questions?