AVEC ’12

Active Safety assessment by Euro NCAP: Electronic Stability

Control and Dynamic Handling

Josep Maria Dalmau, Applus IDIADA Group, China

Sandro Boltshauser, Alexandre Català, Applus IDIADA Group, Spain

Shanghai, 201315, CHINA

Phone: +86 21 6210 0894
Fax: +86 21 5208 0556
E-mail: jmdalmau@idiada.com

Active safety systems such as Electronic Stability Control (ESC) have been clearly demonstrated
to increase vehicle active safety in critical dynamic manoeuvres. With such a high potential for
reducing road accidents, the ESC system is being more widely fitted all over the world. As a result
and in parallel with regulation initiatives, consumer organizations are expanding their assessment
capabilities from passive towards active safety. Since 2008, ESC fitment promotion has been the
first step for worldwide consumer organizations such as ANCAP (Australia), NASVA (Japan) and
Euro NCAP (Europe). However a major step has now been taken by Euro NCAP with the
inclusion of the first active safety test in the overall Euro NCAP safety assessment program.

Since 2011, Euro NCAP performs the so called “Dwell sine” test on all cars that meet the ESC
fitment requirement. This test procedure is based on the European ESC homologation test, part of
the ECE R13H regulation, and reproduces a severe obstacle avoidance manoeuvre where the ESC
efficiency is checked in terms of yaw stability and lateral response. With the current test procedure,
a PASS/FAIL result is obtained considering the applicable thresholds.

Taking the experience from the dynamic tests of 2011 and onwards, Euro NCAP and its members
are continuing to work on possible refinements to the ESC test and assessment methods in order
evolve from PASS/FAIL to a quantifiable and therefore rateable results for ESC assessments.
Therefore additional parameters and test criteria as well as additional manoeuvres have been
analysed in order to develop an assessment scenario covering a wider range of critical driving
situations.

As conclusion, ESC systems on the European market, in general show good results. All vehicles
which have been assessed passed the “Dwell Sine” requirements and show similar results in terms
of safety. However differences can be identified, considering a wider assessment protocol with
additional manoeuvres and criteria.

Topics / Active & Passive Safety Systems

1. INTRODUCTION From the first results presented in 1997, Euro NCAP has
progressively increased the scope and requirements of
Euro NCAP organizes vehicle safety tests providing the safety assessment programme covering: occupant
consumers with a realistic and independent assessment protection, child protection, pedestrian protection and
of the safety performance of some of the most sold cars safety assist. It has been in February 2011, when Euro
in Europe and abroad. NCAP has included the first active safety test as part of
the overall safety assessment programme. The test is
Established in 1997, Euro NCAP is composed of seven known as “dwell sine” and it is performed on all cars
European Governments as well as motoring and that meet the fitment requirements. The points for ESC
consumer organizations across Europe. Euro NCAP has account for the safety assist box and are only awarded
rapidly become a catalyst for encouraging significant when the vehicle passes this test.
safety improvements to new car design.
 AVEC ’12

the safety assessment procedure. In terms of ESC, the

applicable regulation is the ESC homologation
procedure (ECE R13H [1]) dwell sine test procedure,
which originally was developed by NHTSA together
with the industry (Autoalliance) known as FMVSS126
(NHSTA TP-126-02 [2]) and later harmonized into
GTR8.

Based on the ECE R13H, ESC is to be mandatory from

November 2011. With this timeline, Euro NCAP is
currently considering three different phases for the ESC
assessment:

- Phase 1: From February 2011 to end of 2013,

Fig. 1 Scope of Euro NCAP vehicle safety assessment
Euro NCAP performs the sine with dwell test
based on the ECE R13H with PASS/FAIL
The tests are carried out by Euro NCAP accredited test result (Euro NCAP ESC test [3])
facilities that have proven to be capable of delivering - Phase 2: From 2014, Euro NCAP will extend
reproducible results. Applus IDIADA group is an the ESC test and assessment protocols to
official test facility for Euro NCAP ESC testing as well multiple test scenarios
as a member of the Euro NCAP ESC working group
responsible for the definition and validation of the ESC
ESC R13-H Regulation in force
test and assessment procedures. pass/fail

ESC fitment Fitment + Fitment

2. DWELL SINE TEST requirements ECE R13H ECE R13H
Multiple test
scenarios
only test pass/fail test pass/fail

2.1 Introduction 2009-2010 2011 2012-2013 2014

Phase 1 Phase 2
The dwell sine test is an open loop dynamic manoeuvre
that reproduces a closed loop obstacle avoidance (lane Fig. 3 ESC assessment: Euro NCAP vs Homologation
change). This is a safety critical situation where the
driver is suddenly surprised and tries to avoid the As Euro NCAP phase 1 applies the same test procedure
collision applying a quick and large steering wheel input. (Euro NCAP ESC test [3]) and requirements than
In this situation, vehicle’s stability is compromised and homologation (ECE R13H [1]), it can be observed that
can potentially lead to an oversteer (spin-out) situation. Euro NCAP phase 1 advances the homologation
requirement by almost one year. After that, phase 2 will
be active in order to consider the ESC assessment on
additional test scenarios.

2.3 ESC Working Group Activities

The adoption of the dwell sine test as the first active
safety assessment in Euro NCAP, is the result of the
complementary efforts of Euro NCAP secretariat and
the ESC working group.

The ESC working group was the responsible for the

technical validation of the ESC test procedure
corresponding to Euro NCAP phase 1. As the ESC
homologation regulation test (ECE R13H [1]) dwell
sine was already a completely defined and accepted test
procedure, the work for phase 1, was mainly focused on
the validation of the test repeatability over the different
Fig. 2 Closed loop lane-change vs dwell sine (BOSCH)
test laboratories. For this activity a comprehensive
round robin test and analysis process was performed.
With such a specific steering wheel input pattern, it is
necessary to perform this test using a steering robot. In terms of further requirements on phase 1 test, there
This on the other hand, also increases the level of are several issues at the moment in discussion in the
repeatability and objectiveness. ESC working group. As in the previous steps, the
evolution of those current issues will be guided by Euro
2.2 Euro NCAP Approach NCAP’s roadmap over the next months.
Following the same approach used for passive safety
assessment, Euro NCAP takes the applicable
homologation procedure as basis for the development of
 AVEC ’12

The following sections summarize the status of the

current discussions:

Revision of current thresholds

The current pass/fail result is based on the fulfillment of

three objective requirements: two based on stability and
one based on responsiveness.

The thresholds for those test requirements were

determined by comparing the vehicle performance in
ESC on and ESC off configurations. Therefore,
fulfilling those test requirements can be understood as
an objective prove of ESC system fitment.

The following figures illustrate how the current

thresholds were established for both requirements:
stability, by using the yaw rate ratio parameter and
responsiveness, by using the lateral displacement Responsiveness
parameter.

Fig. 5 Lateral displacement threshold definition

(NHTSA)

As these thresholds were established during the

development of the FMVSS126 (NHSTA TP-126-02 [2])
based on the performance of vehicles (and ESC)
fabricated in 2005 or older, the Euro NCAP ESC
working group is considering the possibility to update
them based on the performance of current vehicles (and
ESC). For this reason the working group was collecting
test data of dwell sine tests performed in 2010 (before
the test was official) and now is trying to reproduce the
above shown graphs according to the current level of
vehicle (and ESC) performance. Those graphs might
Stability then be used for setting new pass/fail thresholds for nrxt
phases.
at T0+1  35%
Considerations on steering speed and torque
at T0+1.75  20%
As the dwell sine requires to be performed by a steering
Fig. 4 Yaw rate ratio threshold definition (Autoalliance) robot, little attention is paid to the actual steering speed
and torque that is required in order to execute the test
input. Considering the fact that the dwell sine
manoeuvre reproduces a closed loop lane change,
therefore the test procedure should consider the fact that
a human driver is performing the manoeuvre. This
basically concerns two questions.

The first one is that the test procedure should avoid to

reproduce steering inputs that are clearly out of the
human capabilities in terms of steering speed (such as
700 to 1000 º/s) as they are not representative of a
human driving situation.

The second one concerns a potential new requirement

that could indicate the level of steering effort that needs
to be applied in order to drive the manoeuvre. This
effort is usually not realized as powerful steering robots
are to perform the manoeuvre (torque capacity about
 AVEC ’12

50-60 Nm). However considering the fact that the The value of this parameter can be interpreted as
manoeuvre represents a safety critical situation (obstacle follows:
avoidance), is therefore obvious to require that the
steering wheel effort is also within the human  0 expected yaw response
capabilities.
> 0 over responsive yaw motion
The following diagram illustrates the human capabilities Euro NCAP
EuroDwell
NCAP Sine
Dwell Sine
in terms of steering wheel speed and effort for both an <Driver
0 Intention
under responsive yaw motion
Following vs. Input Factor
yawR ratio at 1 vs. Input Factor

average female and a large male (blue and pink lines). Test Conditions: Test Speed: 80 km/h
Test Conditions: Test Speed: 80 km/h
Analysis Date: 13/07/2010
Analysis
JMD Date: 13/07/2010
The following diagram illustrates how for a three
Analysed by:
Executed by:Analysed
FP by: JMD
The green, red and blue colors correspond to the vehicles meeting the current test requirements,
Executed
Configuration: ESP on by: FP
Configuration: ESP on
evolution of the steering wheel speed and effort applied
Euro NCAP Dwell Sine
additional assessment can be performed by using the
during the dwell sine manoeuvre
Steering-wheel on arate
torque vs. Steering-wheel three different new
Mazda 3“predictability of response”
Citroën DS3 parameter.
Mercedes E-Class
vehicles. TheAnalysis
orange box
Date: 03/03/2011
contains the information that Mazda 3 Citroën DS3 Mercedes E-Class
should not beConfiguration:
considered ESP on as it falls out of human
Weight Condition: Driver Only Weight
1
0.9 10
steering wheel Testrate
Speed: capabilities.
80 km/h Disregarding this part of
0.8 9
the graph it is possible to see how the three vehicles 0.7 8
clearly exceed steering effort capabilities on the red (fail) 0.6
7
area.
Mazda 3 - DS 80 Citroën DS3 - DS 80 Mercedes E-Class - DS 80
0.5
6
0.4
5
Driver Intention Following

0.3
90 4
0.2
SWR out of human capabilities
0.1 3
80 0 2
yawR ratio at 1

-0.1 1

70
FA -0.2 0

60
IL -0.3
-0.4
-1
-2
-0.5
-3
-0.6
-4
50 -0.7
-5
-0.8
-6
-0.9
40 -7
-1
-12-11-10
-8 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12
Input Factor
30 -9
Fig. 7 Predictability of response comparison
-10
-12-11-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12
20 Input Factor
The green area corresponds to the best performance area
where the absolute value of the parameter is low
10
(between 0 and 0.2), therefore very close to the
expected response. The orange area corresponds to the
0 medium level of performance with the absolute values
0 200 400 600 800 1000 1200
Steering-wheel rate [º/s] of the parameter between 0.2 and 0.5 and the red area
corresponds to the poorest performance with absolute
Fig. 6 Human capabilities vs test input
values of the parameter higher than 0.5. So, from the
vehicle samples considered, it is important to highlight
New parameter: predictability of response
the performance of the green vehicle as it is the one
performing closest to the expected performance area
Another important point to be considered in the analysis
(close to 0 value) whereas the poorest performance
of the manoeuvre is that beyond pass/fail discrimination,
correspond to the red vehicle with a clear over
Euro NCAP test procedures usually aims at performance
responsive yaw motion performance over the whole
evaluation. Therefore, analyzing the characteristics of
severity range (input factor).
the manoeuvre, the ESC working group suggested
considering a new parameter named “predictability of
response”. This parameter is an indicator of how
accurately the vehicle is following the driver intention
in terms of yaw response. It is important to highlight
that this parameter is normalized with respect to the test
track and to the vehicle. Therefore the repeatability of
the parameter when testing across different test tracks
can be expected to be within the acceptable levels.
 Euro NCAP Dwell Sine
Driver Intention Following vs. Input Factor

Vehicle: VW Tiguan 2008

Test Speed: 80 km/h
Weight Condition: Driver Only Weight
Temperature: 7ºC
AVEC ’12
Test Date: 13/01/2009 Wind Speed: 2m/s
Analysis Date: 13/07/2010 Configuration: ESP on
The following diagram illustrates another example of
Analysed by: JMD Test Track: PDB
Executed by: FP
analysis of vehicle (and ESC) performance using the
“predictability of response” parameter. In this case, the Additional test scenarios
results correspond to a SUV vehicle. As main objective of phase 2 of Euro NCAP ESC
1 SWA 45º 1.5 assessment, the working group is now discussing the
SWA 60º 2
SWA 75º 2.5 possibility to include additional test scenarios and
SWA 90º 3
SWA 105º 3.5
SWA 120º 4
therefore, test procedures in order to assess the ESC
SWA 135º 4.5
SWA 149º 5
performance on a more complete and representative
0.5 SWA 164º 5.5
SWA 180º 6
range of situations.
SWA 195º 6.5
SWA 210º 7
SWA 225º 7.5
SWA 240º 8
The following table summarizes the additional driving
SWA 254º 8.5
SWA 270º 9
situations and related tests that are considered at the
0
SWA 285º 9.5
SWA -45º -1.5
moment (US stands for understeer and OS stands for
SWA -60º -2
SWA -75º -2.5
oversteer).
SWA -90º -3
SWA -105º -3.5
SWA -120º -4
SWA -135º -4.5
Table 1 Additional test scenarios
SWA -149º -5

-0.5 SWA -164º -5.5

SWA -178º -6
Driving situation Related test Procedure based on
SWA -193º -6.5
SWA -209º -7
SWA -224º -7.5 Avoidance: Step steer +
SWA -239º -8 ISO 7401
SWA -254º -8.5 progressive OS Power off
SWA -268º -9
SWA -283º -9.5
-1
-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12
Closing radius: Steady state
Input Factor ISO 4138
progressive US/OS ramp input
Fig. 8 Predictability of response of SUV

Out of the evolution of the parameter during the SUV Throttle release in a
Power off ISO 9816
test, it is possible to realize a big step down of the value turn: OS
approximately around the medium of the input factor
range. For low input factor test runs, the vehicle Brake application in a
Braking in a turn ISO 7975
performs with a very low parameter value therefore, turn: OS
closely following the expected response. However, from
a certain input factor (steering wheel amplitude and Lane change:
Lane change ISO 3888-2
lateral acceleration), the vehicle performance sharply progressive OS
evolves to a much lower value (around -0.5), therefore
showing a very under responsive yaw motion 3. SUMMARY/CONCLUSIONS
performance.
The adoption of the dwell sine test as part of the overall
In such a SUV vehicle (high center of gravity vehicle), Euro NCAP assessment programme, is the first step of
this clear difference is due to the rollover mitigation consumer organization towards active safety
strategy of the vehicle being activated. The graphs
Euro NCAP Dwell Sine Euro NCAP Dwell Sine
assessment.
Lateral deviation vs. Distance Lateral deviation vs. Distance

below show the vehicle trajectory evolution and how it

Vehicle: VW Tiguan 2008
Test Speed: 80 km/h
Weight Condition: Driver Only Weight
Temperature: 7ºC
Vehicle: VW Tiguan 2008
Test Speed: 80 km/h
Weight Condition: Driver Only Weight
Temperature: 7ºC

correlates to the predictability of response parameter

Test Date: 13/01/2009
Analysis Date: 13/07/2010
Analysed by: JMD
Wind Speed: 2m/s
Configuration: ESP on
Test Track: PDB
Test Date: 13/01/2009
Analysis Date: 13/07/2010
Analysed by: JMD
Wind Speed: 2m/s
Configuration: ESP on
Test Track: PDB The use of state-of-the art test equipment and the strong
evolution. Executed by: FP Executed by: FP

focus on objectiveness and repeatability of the test

20 20
results, enable consumer organisations to expand their
18
16
SWA 45º 1.5

18
SWA 60º 2
16
SWA -45º -1.5

SWA -60º -2
assessment capabilities from passive towards active
14
12
SWA 75º 2.5
14
12
SWA 90º 3
SWA -75º -2.5

SWA -90º -3
active safety.
10 10
SWA 105º 3.5 SWA -105º -3.5

8 8 SWA -120º -4

REFERENCES
SWA 120º 4

6 6
SWA 135º 4.5 SWA -135º -4.5
4 4
Trajectory [m]
Trajectory [m]

SWA 149º 5 SWA -149º -5

2 2
SWA -164º -5.5
0
[1] United Nations, “ECE R13-H, Brake system
SWA 164º 5.5
0
-2 -2
SWA 180º 6 SWA -178º -6

-4
homologation”, 2008.
-4 SWA 195º 6.5 SWA -193º -6.5

-6 -6
SWA 210º 7 SWA -209º -7

-8 -8
-10
SWA 225º 7.5
-10
SWA -224º -7.5

SWA -239º -8
[2] U.S. Department of Transportation,
“LABORATORY TEST PROCEDURE FOR
-12 -12
SWA 240º 8

-14 -14
SWA 254º 8.5 SWA -254º -8.5

FMVSS 126, Electronic Stability Control Systems”,

-16 -16
SWA 270º 9 SWA -268º -9

-18 -18 SWA -283º -9.5

SWA 285º 9.5

-20 -20
-10 0 10 20 30 40 50
Distance [m]
60 70 80 90 100 -10 0 10 20 30 40 50
Distance [m]
60 70 80 90 100
TP-126-02, 2008.
Fig. 9 Trajectory evolution of SUV [3] Euro NCAP, “THE DYNAMIC TEST OF CAR
ELECTRONIC STABILITY CONTROL (ESC)
SYSTEMS PROTOCOL”, version 1.1, December 2010.