Volume 3, Issue 3, March– 2018 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

HIL Testing of ADAS Features

Anju K Vincent, V.Bagyaveereswaran
Control and Automation Department
Vellore Institute of Technology
Vellore, Tamil Nadu

Abstract:- In recent years there is a drastic increase in the ECU is the brain of ADAS system. ECU has a
road accident. This was mainly because of the drivers processor unit, signal conditioning circuits and memory. The
inattention. Thus safety of the passengers and the driver algorithm for the ADAS feature is flashed as a software in
became one of the important aspect of automotive the ECU. The sensor data is processed by the ECU. The sensor
industries. Automotive industries developed features that data acts as input for the algorithm. The decision of the ECU
assist the driver to avoid accidents. These features are depends on the detected data. Accordingly the ECU generates
called Advanced Driver Assistance System(ADAS). These control signal.
features are implemented using sensors and Electronic
Control Unit(ECU). There is need to test and verify the A system becomes reliable only after testing and
ECU to make sure that they perform the ADAS functions. verifying it. The software flashed in the ECU performs the
Hardware-In-Simulation(HIL) testing is used to test the ADAS function. The software should be tested in real time
ADAS ECU. It is a testing technique used to test the ECU environment to make the system more robust. Techniques are
in a real-time environment. developed to test and verify the ADAS system. Hardware In
Loop(HIL) test technique is used to test and verify the ECUs
Keywords:- ADAS, sensors, ECU, testing, HIL. of ADAS system.
I. INTRODUCTION
This paper has four section. Section one is the
The industrial revolution accelerated the growth of introduction. Section two describes the ADAS features.
automotive industries. Advanced technologies embedded in Section three discuss about the test methodology and flow of
the vehicles increased the number vehicles. As number of HIL testing. Section four explain about the hardware setup and
vehicles on roads increased , driving became more software used for HIL testing. Section five gives the results of
challenging. The inattention of the drivers caused many the HIL testing. The paper concludes with section six by
accidents. Thus safety became one of the most important summarizing the whole paper.
aspect of automotive industries. System that had features to
assist the driver were developed to avoid accidents. These II. ADAS FEATURES
features are called Advanced Driver Assistance
System(ADAS). Many accidents occur due to the inattention of the
drivers. Automotive industries started research to develop
The main ADAS features are Lane Departure technologies to prevent accidents. Safety of the
Warning(LDW), Lane Keep Assistance(LKA), Adaptive driver/passengers become one of the most important aspect of
Cruise Control (ACC) and Traffic Sign automotive industries. This led to the development of
Recognition(TSR).These features are developed to avoid technologies that prevented collision of vehicles. Intelligent
accidents. The features are implemented using sensors and assistance system known as Advanced Driver Assistance
ECUs. Radar and camera are the main sensors that gather data System(ADAS), was developed to assist the driver to avoid
for the ADAS system. Radar can detect the distance between the accidents. Some of the ADAS features are:
the target and host. It also detects the speed of the target.
Camera captures the image and video Some of the sensors are A. Lane Departure Warning(LDW)
Mid-Range Radar(MRR), Short Range Radar(SRR). They are Lane Departure Warning system reduces the amount
mounted on different parts of the vehicle. The position at of unintended lane departures. The LDW functionality can be
which the sensors are placed depends on the function it has to divided into two processes:
implement. For parking assistance camera is placed in the rear
part. Whereas for detecting traffic the camera is placed in the  Lane Detection: The lane markings and road edges are
front part of the vehicle. Thus sensors play a vital role in detected then and estimate the car’s position in the lane
implementing ADAS features. Radar can detect the distance  Warning: The above details are used to give a warning to
between the target and host. It also detects the speed of the the driver, if there is an unintentional lane departure.
target. Camera captures the image and video.
Here camera acts as a sensor to detect the lane
marking. The lanes are detected and the signal is given to the

IJISRT18MA397 www.ijisrt.com 594

Volume 3, Issue 3, March– 2018 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
ECU. The control unit process the data. The warning is given
to the driver when the vehicle is about to go out of the lane. Critical traffic scenarios are created to test the ECU.
The above environment is developed by using the DSPACE
B. Lane Keep Assist(LKA) hardware and software tools.
This feature is similar to lane departure warning. The
lane markings and road edges are detected by the camera then IV. TEST SETUP
and estimate the car’s position in the lane. These details are
used to give a warning to the driver, if there is an unintentional The HIL test setup has both hardware and software
lane departure. This feature is similar to lane departure tool. These tools are developed by DSPACE.
warning. The extra function is, if the driver does not take any
action ECU sends signal to gently steer back the vehicle onto A. Hardware setup
the appropriate lane. SCALEXIO is used as a simulator for HIL testing of
ADAS features. Scalexio is developed by DSAPACE.
C. Traffic Sign Recognition(TSR) Scalexio simulate the system that is controlled by the ECU.
Traffic sign board mainly give information about the Here the vehicle is the simulated system, thus Scalexio acts as
speed limits, turns, warning or conveying information. One of a real time vehicle. Scalexio perform the following function.
the main aim of signs is to alert the driver to avoid accidents.
At times drivers does not follow these signs which leads to  It can generate sensor signals that is required for an ECU.
accident. TSR avoids this situation.  It can measure the signal generated by the ECU for
controlling the actuator.
Camera act as sensor and identifies a traffic sign  It connects the signal generated by the ECU to loads.
depending on the color and shape of the board. The  It act as the power supply for the ECU.
information extracted from the boards are displayed in the
dash board of the car and alert the driver to take action B. Software
accordingly. This reduces the occurrence of accidents caused Only once we configure the Scalexio it act as a
by ignoring of traffic signs by the driver.In all the above vehicle. The traffic scenario that is played to the camera
features camera acts as the sensor. We test the ECU that gets should be developed. All these are implemented using
data from the camera. software tools provided by DSPACE. The main software
components are
III. TEST METHODOLOGY
The reliability of any system increases only after testing and  MATLAB
verifying it. The functions that the ECU should perform is  Configuration Desk
decided by the software flashed in the ECU. Thus software  Model Desk
testing is very important. The software is tested in the  Motion Desk
development stage. During the development phase it is not  Control Desk
practical to test the ECU in a vehicle as it demands more time,
cost and it has a risk factor. But if we are able to create a real
time environment to test the ECU, then it is less dangerous,
cost effective and more robust. This type of testing is called
Hardware-In-Loop(HIL) testing. In HIL the entire vehicle
system is modeled. The sensors, actuators, all inputs and
outputs, communication channels that interact with the Device
Under Test (DUT) are modeled.

For testing ADAS features we need to create a traffic

condition similar to real time traffic. Only then we can verify
the ECU function. The traffic environment which is created is Fig 1:- HIL block diagram
played to the camera that detects the data for the ADAS ECU.
The camera takes the video as real time scenario. The camera  MATLAB
detects the data from the video and this data is given as input
to the ADAS ECU. The ECU analyses and process the data. DSPACE provides a basic Automotive Simulation
And the ECU takes decision accordingly. The ECU decision is Model (ASM). This model is accessed through MATLAB.
generated as an output signal. This signal can either be a ASM has the dynamics of a car. The main parts of a care are
warning to the driver or can be a input to other ECU to take Engine, vehicle body, drive train and four wheels.
appropriate action. This can be monitored to verify whether
the ECU performs its function.

IJISRT18MA397 www.ijisrt.com 595

Volume 3, Issue 3, March– 2018 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
The traffic environment is created using the tools
provided by model desk. Using road generator of model desk
road models are generated. Maneuver is developed using
model desk. Maneuver describes how a vehicle should move.
It defines when to start a vehicle, when to change the lane,
when to take a turn, when to brake the vehicle. The fellow
traffic vehicles are developed using model desk.

 Motion Desk
Motion desk is provided by DSPACE to visualize the
Fig 2:- Simulink model movement of the mechanical objects in the 3D world. The
This model includes the suspension kinematics, steering and vehicle model is parameterized in model desk. Roads,
brakes, tire road friction, forces and moments, a drive train. maneuver and traffic objects are created in model desk. But
we cannot visualize the movement of vehicle.
 Configuration Desk
Motion desk is used to visualize the motion of the
The real-time application runs in the Scalexio vehicle along the road and the maneuver. The data in the
hardware. This application is divided into behavior model and model desk is synched with the motion desk through the
I/O model. The behavior mode is developed in the MATLAB. model desk. Only after the synching process we can visualize
It is the model of the controlled system. This model only the model in motion desk.
contains the algorithm of the controlled model. But through
this we cannot access the hardware. The access to the
hardware is implemented through configuration desk. The
function like accessing the hardware, generating I/O signals,
measuring of I/O signals all are implemented through
configuration desk. The behavior model in the MATLAB and
the I/O model should be connected.
This is done through the model interface feature of
configuration desk. The virtual ECUs can also be created in
Configuration desk. Through the built process in
Configuration desk an executable format of the model is
created and this runs as a real-time application on the Fig 4:- MotionDesk
hardware. Now the hardware act as a vehicle.  Control Desk
Control desk is a component of DSPACE through
 Motion Desk which we can perform calibration of an ECU. Through control
The MATLAB has the ASM model. It is desk we can access the parameters of the model. We can
parameterized using model desk. Model desk acts as a control and monitor the parameters of the model. In control
graphical user interface for changing the parameters and desk we can develop an GUI to access and control the
managing the parameters of the ASM models. Motion desk parameters of the model.
also provides tools for creating roads and maneuver.

Fig 5:- Control Desk

Fig 3:- Model Desk

IJISRT18MA397 www.ijisrt.com 596

Volume 3, Issue 3, March– 2018 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Different test cases are developed to test the ADAS
features. These test cases are created and executed using the
above tools.
V. RESULTS
The ADAS ECU is tested in the real-time
environment. The traffic scenarios created and played to the
camera. The Camera detects the data. The ADAS ECU
performs its function. The output action of ADAS ECU is
monitored. It if found that ADAS ECU performs its function
accordingly when it is tested in critical scenarios. The ADAS
ECU is tested and verified in all possible scenarios. The report
is generated.
VI. CONCLUSION
To reduce accidents ADAS features are implemented
in the vehicles. ECUs perform the ADAS functions. It is
important to test and verify the ADAS features. If we test the
ECU in a real-time environment similar to on-vehicle test,
then the test will be more robust and valid. Thus we use HIL
testing. In this testing technique a real-time environment is
developed and the ECU is tested. The main advantage is that
test scenarios that has risk factor while testing on a vehicle can
be tested using HIL testing. This one of the reason for using
Hil testing. This technique can be used for testing most of the
close loop system. The obtained results shows that this is a
reliable testing technique.
REFERENCES
[1]. J. Q. Wang, L. Zhang, D. Z. Zhang, and K. Q. Li, “An
adaptive longitudinal driving assistance system based on
driver characteristics,” IEEE Trans. Intell. Transp. Syst.,
vol. 14, no. 1, pp. 1¡12, Mar. 2013.
[2]. J. Allen, “Implementation of HIL Testing Systems for
Aerospace ECUs,” SAE Technical Paper 2005-01-3385,
Tech. Rep., Oct 2005.
[3]. C. G. Kiran, L. V. Prabhu, R. V. Abdu, and K. Rajeev,
“Traffic sign detection and pattern recognition using
support vector machine,” in Proc. ICAPR, 2009, pp. 87–
90 .
[4]. Broggi, A., Cerri, P., Medici, P., Porta, P. P., & Ghisio,
G.,” Real time road signs recognition”, In IEEE
Intelligent Vehicles Symposium, 2007,pp. 981-986.
[5]. J. M. Cho et al., "Design and implementation of HILS
system for ABS ECU of commercial vehicles," ISIE 2001.
2001 IEEE International Symposium on Industrial
Electronics Proceedings(Cat. No.01TH8570), Pusan,
South Korea, 2001, pp. 1272-1277 vol.2.
[6]. F. Co􀁳kun, O. Tuncer, E. Karshgil, L. Gllven, "Vision
System Based Lane Keeping Assistance Evaluated in a
Hardware-in-the-Loop Simulator," ASME ESDA 2010,
Engineering Systems Design and Analysis Conference,
Istanbul, accepted,2010.
[7]. Aly, Mohamed. "Real time detection of lane markers in
urban streets."Intelligent Vehicles Symposium, 2008 IEEE.
IEEE, 2008.

IJISRT18MA397 www.ijisrt.com 597