Naturalistic Driving – Naturalistic Observation 4 th CEE Round Table 3-4 October 2011, Budapest Veronika Zuser, KFV
0410 08 Veronika Zuser Naturalistic Driving
Oct 12, 2014
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
Naturalistic Driving – Naturalistic Observation
4th CEE Round Table 3-4 October 2011, Budapest Veronika Zuser, KFV
What is Naturalistic Driving?
Under the naturalistic driving (ND) approach, the behaviour of road users is observed unobtrusively in a natural setting for a long period of time.
By installing data acquisition systems (e.g. data logger, cameras) within the vehicle or site-based, driver and driving performance can be observed within their natural surrounding.
2 04.10.2011
ND system
3 04.10.2011
ND in Road Safety work
4
Road Safety work / Accident investigation
reaction based on: statistical data
Accident data management Black Spot Management Network Safety Management In-Depth studies
prevention based on: planning,
expertise, road user behaviour
Road Safety Impact Assessment Road Safety Audit / Inspection Naturalistic Observation
Site-Based-Observation ND / NR
04.10.2011
State of the Art ND - EU
PROLOGUE (2009-2011)
PROmoting real Life Observations for Gaining Understanding of road user behaviour in Europe
Feasibility study for a large scale European naturalistic observational study
Website: www.prologue-eu.eu
ND pilots in AT, NL, GR, ES, IL
9 partners from 7 countries
5 04.10.2011
Added value of Naturalistic Observation
Traditional research: instrumented cars, simulators, police records, in-depth, interviews…:
No natural environment
No direct, systematic observation of a conflict, near miss or crash
6 04.10.2011
Issues for naturalistic observation: the effect of road design characteristics, or weather conditions on the
interaction between driver and vehicle;
driving style comparison of specific road user groups, e.g. novice drivers;
the identification of crash contributing factors;
the prevalence of mobile phone or other in-car information devices and the relationship with particular behaviour patterns or crashes;
the effect of particular interventions, e.g. eco-driving;
the interaction between motorised vehicles and vulnerable road users;
Purposes of Naturalistic Observation
Two different strategies
1. Large database for subsequent analyses of post-hoc research questions - implicit assumptions about research topics
2. Study focusing on a limited number predefined research questions
Choice of strategy has implications for study design
Strategy 1 random sampling to obtain representativity
Strategy 2 targeted sampling guided by research hypotheses
Combining 1 and 2 is recommended
Stratified sample:
Large random sample of general driver population
Over-sampling of specific driver/vehicle categories, regions, etc., based on research questions
7 04.10.2011
Phases in large-scale ND study: FESTA „V“ modified
07.12.2011 8
Imple
me
nta
tio
n p
lan
User/stakeholder
involvement
Research questions and
hypotheses
Measures and sensors
Data acquisition and
decoding
Data analysis
Research questions and
hypotheses analysis
Socio-economic impact
assessment
Dissemination
Topics selection
Ethical and
legal issues
Database
The ND Chain
Measures
Performance
indicators
Performance
indicators
Study/
design
State of the Art ND – USA
100 car ND study (VTTI) (2003/2004)
- 100 participants, 3.2 m vehicle-km, 43,000 hours of data gathered
- findings: distraction/inattentiveness = cause of 80% of all analysed crashes and near crashes
SHRP 2 (TRB) (2011-2014)
US Strategic Highway Research Program
- large scale ND
- 3,100 participants, 12-24 months in 6 regions in the US
9 04.10.2011
State of the Art ND - KFV
DrivEkustik (2011-2013)
Driver behaviour in and acoustic perception of e-cars
- Compare driving behaviour of users of e-cars and cars with combustion engine with ND (3 ND equipped cars, ca. 6 months)
- Interaction VRUs – e-cars in comparison to cars with combustion engine analysis of potential conflict situations
- Measure perceptibility of different kinds of e-vehicles
- Develop recommendations for the safe integration of e-vehicles in traffic systems
04.10.2011
10
04.10.2011
State of the Art ND - KFV
DrivEkustik Data acquisition system: p-drive
11
State of the Art ND - KFV
Site-based observation Vienna (2010- 2011)
Video observation of a pedestrian crossing in Vienna
- Purpose of project - Analysis of driving behaviour at pedestrian crossing
(e.g. speed, willingness to stop,…)
- Collection of scenarios for exposure of persons on pedestrian crossing
- Analysis of critical situations and assessment of interactions between drivers and pedestrians to identify dangerous situations for future enforcement
04.10.2011
12
State of the Art ND - KFV
13
Site-based observation Vienna Data acquisition system: video cameras
04.10.2011
State of the Art ND - KFV
100 car study Austria (2011- 2013)
- 100 ND equipped cars, 12 months
- Urban and rural, mountainous and plain regions in Austria
- Focus of project - road user behaviour of novice and elderly drivers
- road user behaviour in normal conditions, near miss accidents and crashes
- information about underreported accident causes as distraction, inattention and fatigue
04.10.2011
14
Related Documents
