Estimating Potential Safety Benefits for Advanced Vehicle Technologies

Mikio Yanagisawa

The National Transportation Systems Center

Advancing transportation innovation for the public good

U.S. Department of TransportationOffice of the Secretary of TransportationJohn A. Volpe National Transportation Systems Center

June 8, 2016

2

Presentation Outline

Background

How do we project potential safety benefits?

What is the crash problem?

Examine key steps within the process

Projecting safety benefits

3

• Division within the Volpe Center• Research Crash Avoidance: Identify

effective intervention opportunities for vehicle or cooperative based warning and automated systems and estimate potential safety benefits.

– National crash data query and typology– Test procedures and instrumentation– Data mining and analysis of naturalistic

driving data– Safety benefits estimation and simulation

tools

• Also: Safety of Automotive Electronics• Also: Vehicle Cybersecurity

Advanced Vehicle Technology Research

Crash Problem

Definition

Counter-measure Functions

Objective Tests

System Evaluation

Safety Benefits

Estimation

4

Technologies ResearchedLevel Vehicle Feature

DriverDrowsy Driver Detection

Pre-Crash Sensing - Advanced Restraints

Vehicle-Based

Intelligent Cruise Control & Forward Collision WarningLane Change Warning & Lane Drift Warning

Lateral Drift Warning & Curve Speed WarningPedestrian Warning

Cooperative Technology

Intersection Movement AssistLeft Turn Assist

Blind Spot WarningElectronic Emergency Brake Lighting

Do Not Pass WarningVehicle-to-Infrastructure

Vehicle-to-Pedestrian

AutomaticControls

Crash Imminent BrakingLane Keeping Technology

Cooperative Cruise Control

5

Projecting Potential Safety Benefits

Exposure Ratio ≡ Probability of encountering a driving conflictCrash Prevention Ratio ≡ Probability of a crash given an encounter with a driving conflict

𝑆𝑆𝑆𝑆𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶=1 − 𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬 𝑹𝑹𝑹𝑹𝑹𝑹𝑹𝑹𝑬𝑬 × 𝑪𝑪𝑬𝑬𝑹𝑹𝑬𝑬𝑪𝑪 𝑷𝑷𝑬𝑬𝑬𝑬𝑷𝑷𝑬𝑬𝑷𝑷𝑹𝑹𝑹𝑹𝑬𝑬𝑷𝑷 𝑹𝑹𝑹𝑹𝑹𝑹𝑹𝑹𝑬𝑬

𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐶𝐶𝐴𝐴 =# 𝑇𝑇𝐶𝐶𝐶𝐶𝑇𝑇𝐶𝐶𝑇𝑇 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 × 𝑆𝑆𝑆𝑆𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶

• Ratios are estimated from driver/vehicle/system performance data with and without automated vehicle functions

• Approach is used in vehicle-based, vehicle-to-vehicle, and pedestrian safety system research

• Potential to estimate injury mitigation

• Identify and define a safety system

6

Safety Benefits Estimation Data Flow

Safety Benefits

Crash DataPre-Crash Scenarios

Field Data Driving

ConflictsModeling

Crash Probability

7

National Crash Trends

-

1,000

2,000

3,000

4,000

5,000

6,000

7,000

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Num

ber o

f Cra

shes

(T

hous

ands

)

Calendar Year

Injury Crashes Fatal Crashes Property Damage Only Crashes

In 2014: 3,026B Miles 275M Registered 214M LicensedSince 2001: VMT ↑8% Vehicles ↑24% Drivers ↑12%

Source: NHTSA Traffic Safety Facts 2014, DOT HS 512 261

8

Crash Fatalities Trends

0%

2%

4%

6%

8%

10%

12%

14%

16%

0

5,000

10,000

15,000

20,000

25,000

30,000

35,000

40,000

45,000

50,000

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

% o

f All

Fata

litie

s

Num

ber o

f Fat

aliti

es

FARS Crash Year

Total Fatalities % Pedestrians % Cyclists % Motorcyclist

Total fatalities have decreased by 9,521 (↓ 23%)Since 2001: Pedestrians ↑3% Cyclists ↑ 1% Motorcyclists ↑ 7%

Source: NHTSA Traffic Safety Facts 2014, DOT HS 512 261

9

Defining 37 Pre-Crash ScenariosCrash Type Pre-Crash Scenario Crash Type Pre-Crash Scenario

Animal/maneuver No Driver No driver presentAnimal/no maneuver Non-Collision Non-collision - No Impact

Backing Backing into vehicle Object/maneuverControl loss/vehicle action Object/no maneuverControl loss/no vehicle action Opposite direction/maneuverTurn right @ signal Opposite direction/no maneuverStraight crossing paths @ non signal Other - Opposite DirectionTurn @ non signal Other OtherOther - Turn Across Path Parking Parking/same directionOther - Turn Into Path Pedestrian/maneuverOther - Straight Paths Pedestrian/no maneuverRunning red light Rear-end/striking maneuverRunning stop sign Rear-end/lead vehicle acceleratingCyclist/maneuver Rear-end/lead vehicle moving @ constant speedCyclist/no maneuver Rear-end/lead vehicle deceleratingEvasive maneuver/maneuver Rear-end/lead vehicle stoppedEvasive maneuver/no maneuver Other - Rear-End

Hit and Run Hit and run Road edge departure/maneuverTurning/same direction Road edge departure/no maneuverChanging lanes/same direction Road edge departure/backingDrifting/same direction Rollover Rollover LTAP/OD @ signal Sideswipe Other - Sideswipe

LTAP/OD @ non signal Vehicle Failure Vehicle failure

Animal

Control Loss

Crossing Paths

Cyclist

Evasive

Rear-End

Road DepartureLane Change

Left Turn Across Path/ Opposite Direction (LTAP/OD)

Object

Opposite Direction

Pedestrian

Source: Pre-Crash Scenario Typology for Crash Avoidance Research, 2007 NHTSA , DOT HS 810 767

10

Example Pre-Crash Scenarios

Rear-End – Lead Vehicle Stopped

Lane Change

Straight Crossing Paths

Left Turn Across Path / Opposite Direction

11

Crash Prevention Ratio

Crash Probability EstimationField

Operational Tests

Safety ImpactMethodology

Tool

Objective Tests

Historical Research

SIMULATION• Treatment• Crash Counts• Impact Speeds• ΔV Values

Analysis and

Results

INPUTS• Pre-Crash Data• System Data• Driver Data

ANALYSIS• Crash Avoidance• System Effectiveness• Safety Benefits

National Crash

Databases

Exposure Ratio

12

Potential Crash Avoidance Effectiveness

Source: Various publications including: New Car Assessment Program, Notice For Proposed Rulemaking, Insurance Institute for Highway Safety research, and Enhanced Safety of Vehicle research

0%

10%

20%

30%

40%

50%

60%

70%Fo

rwar

d Co

llisio

nW

arni

ng

Inte

rsec

tion

Mov

emen

tAs

sist

Left

Tur

n As

sist

Road

Dep

artu

reCr

ash

War

ning

Adap

tive

Crui

se C

ontr

ol

Elec

tron

ic S

tabi

lity

Pede

stria

n Cr

ash

Avoi

danc

e/M

itiga

tion

Igni

tion

Inte

rlock

Pote

ntia

l Sys

tem

Effe

ctiv

enes

s

Vehicle Feature

13

Example of Potential Safety Benefits

• Deployment, penetration rates

• Driver interaction • Acceptance, usage,

misuse, negligence, and abuse

• False activation• Unintended consequences• Operational boundaries

• Speed, environment

• Crash statistics over time• Improvement of technology

Other Factors

Source: NHTSA V2V Readiness Document, 2014, DOT HS 812 014

-

100

200

300

400

500

600

700

800

IntersectionMovement Assist

Left Turn Assist

Annu

al N

umbe

r of C

rash

es(T

hous

ands

)

Communication-Based Warning System

Crashes Reduced Remaining Crashes

48%

49%

14

Mikio YanagisawaAdvanced Vehicle Technologymikio.yanagisawa@dot.gov

(617) 494 – 3846

Volpe Center55 Broadway

Cambridge, MA 02142www.volpe.dot.gov

Questions and Contact