Naturalistic Research on Powered Two-Wheelers. Martin Winkelbauer (KFV) Martin Donabauer (KFV) Alexander Pommer (KFV) Reinier Jansen (SWOV)

Transcription

1 Naturalistic Research on Powered Two-Wheelers Martin Winkelbauer (KFV) Martin Donabauer (KFV) Alexander Pommer (KFV) Reinier Jansen (SWOV) UDRIVE Webinar

2 Two worlds two populations 2

3 Typical Riding Purposes 75% leisure riders 25% commuters Hardly any overlaps (Austria, 2012, n=1038) Leisure riders Commuters Group riding Travelling Sport Riding Track&Offroad Returning riders Permanent riders Returning riders Permanent riders Returning riders Permanent riders Returning riders Permanent riders Returning riders Permanent riders Returning riders Permanent riders 3

4 Camera positions Forward cameras Feet camera Face camera Driver s action camera Passenger compertment camera Right blind spot camera Rear View

5 Camera Position PTW Forward cameras Face camera Side cameras Top case UDRIVE Webinar

6 DAS overview Cars Trucks PTW Cars Trucks PTW Cars Trucks Cars Cars Trucks PTW Cars Trucks PTW Cars Trucks UDRIVE Webinar

7 UDRIVE PTW Piaggio Liberty 125 8

8 UDRIVE Webinar

9 Research questions particular in PTW Everyday riding 50 km/h right turn, left turn Acceleration from stop etc Safety Critical Event (SCE) detection Test triggers Validate by video Time Headway Read endings, 62% drivers at fault Car data only Use mobileye 10

10 Preliminary results: Time Headway 10% PTW crashes rear ending 62% car at fault research based on car data using Mobileye queries direction on SGL database avoid traffic jam v > 30 km/h v > 0,5 speed limit sidewards distance < 3 m lead vehicle present > 10 s 134 mio records i.e h 22% behind car 1.25% behind truck 0.07% behind PTW for v > 85 km/h distance detection probably not exact enough (currently too few data) 11

11 Frequency (%) Frequency of distance, v < 55 km/h 7,00% 6,00% 5,00% 4,00% 3,00% 2,00% 1,00% 0,00% 0 0,5 1 1,5 2 2,5 3 3,5 4 Distance (s) Bike Car Truck 12

12 Frequency (%) Frequency of distance, 55 < v < 85 km/h 9,00% 8,00% 7,00% 6,00% 5,00% 4,00% 3,00% 2,00% 1,00% 0,00% 0 0,5 1 1,5 2 2,5 3 3,5 4 Distance (s) Bike Car Truck 13

13 Average of distance 1.1 s behind cars 1.2 s behind PTWs 0.9 s behind trucks Explanation for rear endings? Back to conspicuity? 14

14 Everyday riding: Setup Aim: To detect, understand, and possibly prevent motorscooter crashes Approach: Descriptives on everyday riding at urban intersections Measures: Speed choice & g-forces Depending on: Scenarios: Flow, Full stop Manoeuvres: Left turn, Right turn, Straight ahead Driver personalities based on questionnaires 15

15 Everyday riding: Expected results Scenario Speed (km/h) distribution, %above limit Acceleration (g) distribution Pre-stop Pre-man Man Post-man Pre-stop Pre-man Man Post-man Flow Full stop Left X X X X X X X Right X X X X X X X Straight X X X X X X X Left X X X X X X X X Right X X X X X X X X Straight X X X X X X X X 16

16 Everyday riding: Expected results Scenario Speed (km/h) mean,min,max, %above limit Acceleration (g) mean,min,max Full stop Pre-stop Pre-man Man Post-man Pre-stop Pre-man Man Post-man Gender Age Experience Personality M X X X X X X X X F X X X X X X X X Young X X X X X X X X Old X X X X X X X X Novice X X X X X X X X Exp. X X X X X X X X Cat 1 X X X X X X X X Cat 2 X X X X X X X X Cat 3 X X X X X X X X Cat 4 X X X X X X X X 17

17 SCEs: What we're looking for... Recording Vehicle manoeuvres: e.g. speed, acceleration/deceleration, direction, high jerk Driver/rider behaviour: e.g. eye, head and hand manoeuvres External conditions: e.g. road, traffic and weather characteristics 18

18 acceleration (g) Preliminary results: SCEs 19 / 40 scooters 500 hours of riding data Acceleration x / y / x Rotation speed x / y / z y longitudinal x lateral z vertical corrected by average filtered 30 to 2 Hz cut off at 55km/h map-matched GPS speed original y acceleration filtered y acceleration time (s) 19

19 acceleration (g) Preliminary results: SCEs 19 / 40 scooters 500 hours of riding data Acceleration x / y / x Rotation speed x / y / z y longitudinal x lateral z vertical corrected by average filtered 30 to 2 Hz cut off at 55km/h map-matched GPS speed original y acceleration filtered y acceleration time (s) 20

20 -0,94-0,86-0,79-0,72-0,67-0,63-0,58-0,53-0,49-0,45-0,41-0,37-0,33-0,29-0,25-0,21-0,17-0,13-0,09-0,05-0,01 0,03 0,07 0,11 0,15 0,19 0,23 0,27 0,31 0,35 0,39 0,43 0,47 0,51 0,55 0,59 0,63 0,67 0,72 0,77 0,81 0,86 0,93 1 Distribution of longitudinal acceleration Trigger = 0,5 g N

21 -1,72-1,44-1,39-1,27-1,19-1,13-1,06-1,01-0,95-0,9-0,85-0,77-0,71-0,62-0,55-0,49-0,42-0,37-0,32-0,27-0,22-0,17-0,12-0,07-0,02 0,03 0,08 0,13 0,18 0,23 0,28 0,33 0,38 0,43 0,48 0,54 0,6 0,65 0,71 0,77 0,82 0,89 0,96 1,01 1,06 1,15 1,23 1,3 1,42 1,49 Distribution of lateral acceleration Trigger = 0,25 g N

22 Distribution of vertical acceleration Trigger = 0,25 g N

23 Observations at outliers but nothing dangerous obs subjective assessment 87 no reason recognisable 47 in garage 32 curve 29 brake 16 start from traffic light 10 strong brake at zebra (no ped.) 6 gravel road 5 lane change 5 brake for pedestrian on zebra 4 speed hump 4 probably curve (unsecure detection) 4 rough road 3 start (other) 3 strong braking at traffic light 2 start from parking 2 swerve 2 turn 1 start from traffic light and change lane 1 enter parking lot 1 accelerate 1 non-critical interaction with pedestrian on zebra 1 strong braking in congestion 1 brakíng, curve 1 strong braking behind other PTW 1 overtaking bicycle 1 strong braking for parking space 1 strong braking 271 Total

24 Distributions of rotation speed x y 109 cases as acceleration. and a lot of roundabouts and some u-turns z 25

25 NR is not easy, but worth it TRA Conference Paris 2014 Martin Winkelbauer, KFV Phone: martin.winkelbauer@kfv.at /7/2017