2.07 - Innovation Lab - VIAS - MOVIT

20 views
0 Likes
0 0

Share on Social Networks

Share Link

Use permanent link to share in social media

Share with a friend

Please login to send this presentation by email!

Embed in your website

Select page to start with

2. Introduction 26 June 2019 / Slide 2

7. Methodology 26 June 2019 / Slide 7

11. 26 June 2019 / Slide 11

12. Results 26 June 2019 / Slide 12

5. Research questions 26 June 2019 / Slide 5

23. Discussion & Conclusion 26 June 2019 / Slide 23

27. 26 June 2019 / Slide 27 Thank you ! peter.silverans@vias.be sofie.boets@vias.be

22. Research question 2: Do dual drivers perform better in prediction of potential conflicts with PTW’s? Dual drivers • gaze more at conflicting motorcycles • make more glances in their rear view mirror when overtaking an obstacle 26 June 2019 / Slide 22

14. Research question 1: Do dual drivers use different visual strategies than “regular” car drivers? • How long (dwell times) and how frequent (glance frequencies) do they look at certain areas of interest? • How do they visually scan their environment? 26 June 2019 / Slide 14

19. Research question 2: Do dual drivers perform better in prediction of potential conflicts with PTW’s? • How long (dwell times) and how frequent (glance frequencies) do they look at the conflicting object (car or motorcycle)? • Visual inspection of mirrors • glance frequencies and dwell times at left and rear view mirror 26 June 2019 / Slide 19

1. Detection of motorcyclists by car drivers with and without motorcycling experience. A virtual reality driving simulator study including eye - tracking (MOVIT) Sofie Boets, Charlotte Desmet, Daniela Knowles, Alexander Pommer & Susanne Kaiser / peter.Silverans@vias.be Vias institute (BRSI) and KFV Macq Innovation Lab, Brussels, June 14th 2019

6. ▸ Do dual drivers use different visual strategies than “regular” car drivers? • How long and how frequent do they look at certain areas of interest? • How do they visually scan their environment? ▸ Do dual drivers perform better in prediction of potential conflicts with PTW’s? Gaze behaviour in specific situations based on typical accident configurations. • How long and how frequent do they look at the conflicting object? • Do they inspect their mirrors differently? 26 June 2019 / Slide 6

13. ▸ participants ▸ areas of interest 26 June 2019 / Slide 13 31 33 left side mirror motorcycle car speedometer sign pedestrian right side mirror rear view mirror

18. Research question 1: Do dual drivers use different visual strategies than “regular” car drivers? Dual drivers • have a more dispersed vision on the X - axis • have a smaller visual spread on the Y - axis 26 June 2019 / Slide 18

3. ▸ motorcyclists among most vulnerable road user group ▸ opponent in most accidents is a passenger car ▸ most accidents occur in urban areas ▸ driving and impact speeds of the PTW’s are mostly below 50km/h ▸ principle cause for accidents is human error and mainly a failure to see the PTW (in time) Van Elslande , P., J. - Y. Fournier, and C. Parraud , Powered two - wheelers in urban environment: a detailed accident analysis. International Journal of Safety and Security Engineering, 2015. 5 : p. 322 - 335 3. 26 June 2019 / Slide 3

8. ▸ participants : male drivers aged 22 to 54 ▸ materials 26 June 2019 / Slide 8 # 39 min 4 years driving license category B # 39 driving licence category B + category A driving license min 2000 km in the last year (> 125 cc) Car drivers Dual drivers

25. Discussion ▸ not all results reach significance at the p < .05 level ▸ results all go in the same direction ▸ VR for road safety research? ▸ novel tool: eye - tracking + VR ▸ no available software for the eye - tracking data linked to the virtual world (e.g. for automated AOI analysis) ▸ unity programming of VR scenarios proved to be very labor - intensive → limited driver dependent changes in scenarios (constant speed was required) ▸ simulated hands on the steering wheel were distracting ▸ cyber - sickness 26 June 2019 / Slide 25

4. ▸ are better in visually detecting the motorcyclist ▸ have safer driving responses at junctions ▸ are better in hazard perception ▸ and have better search strategies Crundall , D., E. Crundall , D. Clarke, and A. Shahar , Why do car drivers fail to give way to motorcycles at t - junctions ? Accident Analysis and Prevention, 2012. 44 : p. 88 - 96. Hosking, S. G., C.C. Liu, and M. Bayly , The visual search patterns and hazard responses of experienced and inexperienced motorcycle riders . Accident Analysis and Prevention, 2010. 42 : pp. 196 - 202.e.g., 5,6]. 26 June 2019 / Slide 4 Dual drivers

17. How do they visually scan their environment? 26 June 2019 / Slide 17 Y X - 2 ( - 180 ° ) 1 (90 ° ) 2 (180 ° ) - 1 - (90 ° ) 1 (90 ° ) - 1 ( - 90 ° )

10. ▸ obstacle ▸ conflicting motorcycle (2) ▸ control event (2) ▸ stream of cars ▸ conflicting motorcycle (2) ▸ conflicting car (1) ▸ other conflicting and non - conflicting events ▸ crossing pedestrian , car that leaves parking spot,.. 26 June 2019 / Slide 10

15. How long and how frequent do dual drivers look at certain areas of interest? ▸ Dual drivers make more glances at cars than car drivers. 26 June 2019 / Slide 15 2,77 3,05 0,00 0,50 1,00 1,50 2,00 2,50 3,00 3,50 1 2 Number of glances per minute glance frequency at cars * Significance : ** p≤.01; * p≤.05; * p ≤..10 Mann - Whitney U - test

21. Visual inspection of mirrors ▸ Dual drivers make more glances in the rear view mirror . 26 June 2019 / Slide 21 0,34 0,58 0,00 0,10 0,20 0,30 0,40 0,50 0,60 0,70 0,80 1 2 Number of glances per minute glance frequency at rear view mirror * Significance : ** p≤.01; * p≤.05; * p ≤..10 Mann - Whitney U - test

9. ▸ task ▸ follow destination ‘Vienna’ ▸ follow traffic rules and speed indications (50 km/h) ▸ familiarisation ride + 5 scenarios of 2.5 minutes (~2 - 2.5km) ▸ experimental events ▸ intersections ▸ conflicting motorcycle (2) ▸ control event (1) ▸ conflicting motorcycle (2) ▸ conflicting car (2) ▸ control event (2) 26 June 2019 / Slide 9

26. Conclusion ▸ Differences between dual drivers and car drivers ▸ Dual drivers have different visual strategies and are better in predicting potential conflicting events ▸ Dual drivers are constantly aware that a motorcycle might approach and they know which manoeuvres can be expected from motorcycles. 1 ▸ Improving motorcycle detection → teaching car drivers from where and in what circumstances a motorcycle may appear ▸ VR with integrated eye - tracking is a promising and useful tool for future research! 1 Gershon, P., Ben - Asher, N., and D. Shinar, Attention and search conspicuity as a function of their visual context. Accident analysis and prevention. Accident Analysis and Prevention, 2012. 44: p. 97 - 103. 26 June 2019 / Slide 26

16. ▸ Dual drivers spend less time looking at their speedometer and more time looking at pedestrians compared with car drivers. 26 June 2019 / Slide 16 5,10 3,86 0,00 1,00 2,00 3,00 4,00 5,00 6,00 7,00 1 2 Percentage dwell time dwell times at speedometer * 1,41 1,72 0,00 0,50 1,00 1,50 2,00 2,50 1 2 Percentage dwell time dwell times at pedestrians * Significance : ** p≤.01; * p≤.05; * p ≤..10 Mann - Whitney U - test

20. How long and how frequent do dual drivers look at the conflicting object? ▸ Dual drivers make more gazes at motorcycles than car drivers 26 June 2019 / Slide 20 0,74 0,54 0,86 0,62 0 0,2 0,4 0,6 0,8 1 1,2 1 2 Number of glances per minute glance frequency car driver dual driver * ns linear mixed model to estimate the effect of GROUP ( dual driver versus car driver) and OBJECT ( car / motorcycle ) on glance frequencies (effect for motorcycles: estimate = - 0.16, z = - 1.64 , p = .10) Significance : ** p≤.01; * p≤.05; * p ≤..10

24. Summary of the results 26 June 2019 / Slide 24 • more glances at other cars • longer glances at pedestrians • more dispersed horizontal scanning pattern • more glances at the conflicting motorcycle • more glances in the rear view mirror when overtaking an obstacle • longer glances inside the car ( speedometer ) • more dispersed vertical scanning pattern → Dual drivers have different visual strategies than car drivers → Dual drivers are better in predicting potential conflicts with PTW’s

Views

  • 20 Total Views
  • 11 Website Views
  • 9 Embedded Views

Actions

  • 0 Social Shares
  • 0 Likes
  • 0 Dislikes
  • 0 Comments

Share count

  • 0 Facebook
  • 0 Twitter
  • 0 LinkedIn
  • 0 Google+

Embeds 2

  • 2 odoo.macq.eu
  • 9 www.macq.eu