#4 recurrent measurements of user experience in avsauvsilink.org/avs2018/posters/victor...

• Less eyes on road with increased AD experience • Perceived safety increased and then stabilized with increased AD experience Intro to study, car, equipment Lap 1: Manual driving Lap 2: Manual driving 90 min Ca 6 min/lap Intro to AD and demo Lap 3: AD available Lap 4: AD available Lap 5: AD available Lap 6: AD available Lap 7: AD available Post quest. Interview Manual driving to garage Roadworks Intersection Roadworks Slow moving Roadworks Oncoming Roadworks Intersection Roadworks Overtaking Roadworks Slow moving Roadworks Oncoming #4 Recurrent Measurements of User Experience in AVs Victor Malmsten Lundgren, Jonas Andersson, Emma Asker, Azra Habibovic, Maria Klingegård, David Lindström, Alex Voronov - A Method Development Experiment at AstaZero Proving Grounds Research Instutes of Sweden, RISE Viktoria Background Method Results and conclusions Future work Acknowledgements Funded by SAFER Open Research at AstaZero, Knowledge Founda- on, and Strategic Vehicle Research and Innovaon (FFI). The studies connue in real traffic within the project Trust in Intelligent Cars (TIC). The experiment used quantave and qualitave methods to invesgate behavior over me when experiencing highly automated driving (AD). Research quesons: • How does user behavior differ between two occasions? • What factors indicate a shiﬅ from first encounter behavior (tense) to more experienced use (calm)? • Two occasions à 90 min for each test parcipant • Within subject design • Highly automated test vehicle (corr. SAE level 4) • 8 test parcipants (4 female, 4 male ) • 80 km on rural road at Asta Zero proving grounds in Sweden Contact: [email protected] • Variaons in driver behavior and first encounter effects were aenuated over me: • 1st occasion: exhilarated (”fun”, ”cool”, ”impressive”) • 2nd occasion: neutral state (”posive”, ”safe”, ”good”, “calm”) • Several factors indicated the shiﬅ from tense to relaxed, e.g.: perceived safety, glance behavior, me to engage in non-driving tasks • Faster to engage in non-driving tasks with increased AD experience Fig. Data collecon methods and devices Fig. Experiment meline.

Upload: others

Post on 06-Oct-2020

3 views

Category:

Documents

0 download

Report

Download

Embed Size (px):

TRANSCRIPT

Page 1: #4 Recurrent Measurements of User Experience in AVsauvsilink.org/AVS2018/Posters/Victor Malmsten... · • 1st occasion: exhilarated (”fun”, ”cool”, ”impressive”) •

• Less eyes on road with increased AD experience

• Perceived safety increased and then stabilized with increased AD experience

Intro to study, car, equipment

Lap 1:Manual driving

Lap 2:Manual driving

90 min Ca 6 min/lap

Intro to AD and demo

Lap 3:AD

available

Lap 4:AD

available

Lap 5:AD

available

Lap 6:AD

available

Lap 7:AD

available

Post quest.Interview

Manual driving

to garage

RoadworksIntersection

RoadworksSlow moving

RoadworksOncoming

RoadworksIntersection

RoadworksOvertaking

RoadworksSlow moving

RoadworksOncoming

#4 Recurrent Measurements of User Experience in AVs

Victor Malmsten Lundgren, Jonas Andersson, Emma Asker, Azra Habibovic, Maria Klingegård, David Lindström, Alex Voronov

- A Method Development Experiment at AstaZero Proving Grounds

Research Institutes of Sweden, RISE Viktoria Background

Method

Results and conclusions

Future work AcknowledgementsFunded by SAFER Open Research at AstaZero, Knowledge Founda-tion, and Strategic Vehicle Research and Innovation (FFI).

The studies continue in real traffic within the project Trust in Intelligent Cars (TIC).

The experiment used quantitative and qualitative methods to investigate behavior over time when experiencing highly automated driving (AD).Research questions:• How does user behavior differ between two occasions?• What factors indicate a shift from first encounter behavior (tense) to more experienced use (calm)?

• Two occasions à 90 min for each test participant• Within subject design• Highly automated test vehicle (corr. SAE level 4)• 8 test participants (4 female, 4 male )• 80 km on rural road at Asta Zero proving grounds in Sweden

Contact: [email protected]

• Variations in driver behavior and first encounter effects were attenuated over time:• 1st occasion: exhilarated (”fun”, ”cool”, ”impressive”) • 2nd occasion: neutral state (”positive”, ”safe”, ”good”, “calm”)

• Several factors indicated the shift from tense to relaxed, e.g.: perceived safety, glance behavior, time to engage in non-driving tasks

• Faster to engage in non-driving tasks with increased AD experience

Fig. Data collection methods and devices

Fig. Experiment timeline.

U.S. DOT Automated Vehicle Research Activities02f09e7.netsolhost.com/AVS2018/Plenary/1100-1130_Thu_Fulton_an… · Cooperative Automated Vehicle Testing with VDOT and Transurban Objective:

Robyn N. Malmsten, B.A. James R. Evans, M.A. Louisiana ... Stress... · Programs (Donald Meichenbaum & Roy ... Robyn Malmsten: [email protected] James Evans: [email protected] (318)

SIP-adus Field Operational Tests and Regulatory Issues02f09e7.netsolhost.com/AVS2018/Plenary/915-930_Thu_Amano.pdf- Human Machine Interface (HMI) - Cyber Security - Pedestrian Accident

Malmsten företaget med vattenvana

UNDERGRADUATE HANDBOOK - physics.utah.edu program is one of the most rigorous at the University and our students and faculty are exhilarated by the difficult challenge of figuring

MAX PLATFORM FOR AUTONOMOUS BEHAVIORSauvsilink.org/AVS2018/ESS/David+Hofert_The+Platform... · ARTIFICIAL INTELLIGENCE LOW-LEVEL SIMPLE BY DESIGN NARROW SCOPE FOR CONTROL ... Intersection

LOGISTIIKKA SELVITYS 2018...LOGISTIIKKA- SELVITYS 2018 Tomi Solakivi, Lauri Ojala, Sini Laari, Harri Lorentz, Tuomas Kiiski, Juuso Töyli, Jarmo Malmsten, Anu Bask, Oskari Rintala,

Design med hög igenkänningsfaktor324531/FULLTEXT01.pdfCarl Malmsten - Furniture Studies Design med hög igenkänningsfaktor Hur påverkar den oss och vad är typiskt för detta sättet