a review of automated emergency braking system and the ... … · “automatic emergency braking...

21 October 2013 Melaka MARA Professional College & MIROS PC3, Road Transport Academy (APJ), Melaka, Malaysia

Proceedings of the

Southeast Asia Safer Mobility Symposium 2013

Page 38

SAEM 2013-010

A Review of Automated Emergency Braking System and the Trending for Future Vehicles

Likun Xia1 *, Tran Duc Chung2, Khairil Anwar Bin Abu Kassim3

1,2Electrical and Electronic Engineering Department Universiti Teknologi Petronas

31750, Tronoh, Perak, Malaysia

3Vehicle Safety and Biomechanics Research Centre Malaysian Institute of Road Safety Research

*Corresponding author’s email: [email protected]

Abstract - The paper presents review of automated emergency braking (AEB) systems, its current state of the art and the trend for future development. Several important safety measures of AEB systems are time to collision, stopping distance and speed reduction. There is strong need to develop low-cost AEB system to broaden the use of such safety system on various vehicle ranges. Some considerations for the development of AEB systems are suggested to be taken into consideration by dominant automakers in Malaysia namely Proton & Perodua.

Keywords - collision warning (CW) system, time to collision, stopping distance, speed reduction, AEB system

I. INTRODUCTION From 1995 to 2010, on average, each year

289,992 road accidents take place resulting in 6,191 deaths, 39,828 injuries and economic loss through 521,178 vehicles involving in accidents in Malaysia [1]. Malaysia has significantly high annual rainfall and rain days, standing at 2500 mm 178 days respectively [2]. Among several causes of accidents, adverse weather is considered as a main cause [3]. 9.18 % of accidents are caused by adverse weather and up to 7.12 % of accidents are caused by rainy weather [2]. Therefore, it is important to develop an AEB system to assists drivers to avoid or reduce fatal accidents. Since early 2000s [3], AEB systems have been developed by many automakers to assist drivers in braking maneuver in emergency cases. Commercialized AEB models do exist in the global market and have been integrated mainly into high-end vehicles [3, 4]. An example of AEB system working at low speed, less than 35 kph, is Volvo “collision warning with full auto brake and pedestrian detection” (CWAB-PD) [5]. The

system could completely avoid collision with other vehicles. At higher speed higher than 35 kph, the system can only reduce severity of impact [5]. Currently, in Malaysia, there is no low-cost AEB system in the market yet, or the system is not mature enough to be used in rainy weather. Additionally, as part of AEB systems, CW systems working under rainy condition are not yet focused in recent researches. Therefore, low-cost AEB systems need to be developed for developing countries like Malaysia to improve road safety.

II. LITERATURE REVIEW A. Car Accidents Statistics and the Need to Develop AEB Systems

75 % of car accidents are at low speed, less than 32 kph [6]. There are three main causes leading to car accidents: (i) careless or fatigued drivers; (ii) sudden change in road direction at “sharp blends and angles”; (iii) bad weathers (heavy rain, dusty road, and dense fog), slippery roads, natural disasters (earthquake, tsunami) [3]. The third cause contributes to about 19.42 % of car accidents [7]. Compensation for traffic accidents, including car accidents, is very costly in terms of traffic delay hours leading to increase of fuel waste and “congestion cost”: 4.2 billion hours, 2.9 billion gallons and $80 billion in US in 2007 respectively [3]. Most of the accidents are caused by human errors such as less attention on road while driving, inefficient brake force creating not enough deceleration rate (21% has no deceleration, 73% decelerates less than 5 m/s2 [8]). Therefore AEB system needs to be developed to assist driver with braking maneuver to avoid collision, and reduce major and fatal injuries to passengers, collision


Proceedings of the


Page 39

SAEM 2013-010

aftereffect, and traffic congestion. Many automobile makers have introduced such systems to their high-end models since 2003 [3] however the systems are not perfect and need further enhancement to be more effective. B. Basic Components of AEB System

AEB system, commonly named as “automatic emergency braking system” or “autonomous emergency braking system” [5, 8, 9], consists of:

a set of sensors installed on a host vehicle measuring: its velocity, acceleration, distances to other movable or stationary objects e.g. human, other vehicles, big rocks;

a brake control mechanism: to control the vehicle’s brake automatically by electronic controller(s);

controller(s): to process signals fed from the sensors, calculate and estimate potential of collision with the detected objects, provide warning signals to driver and assist driver in reducing vehicle’s speed to avoid collision or mitigate collision impact.

Controller is considered as the most important part and plays critical role in AEB system.

Different manufacturers develop different

AEB systems and currently there is no common standard for AEB systems yet. Table 1 lists several AEB systems and Pre-AEB systems available in the market. The term “Pre-AEB systems” refers to the systems such as Lane Departure Warning System of Peugeot. It provides warning signals to driver when certain potential danger exists. This type of system could be expected to be further developed to become AEB system in the future.

Table 1: Several AEB Systems & Supplementary

Technologies [10, 11]

Brand AEB Systems & Supplementary Technologies

Audi Pre-Sense Basic Pre-Sense Front Pre-Sense Front Plus Adaptive Cruise Control (ACC) with Stop & Go function Audi Side Assist + Pre-Sense Rear Active Lane Assist Night Vision Assistant

BMW iBrake 3 Driving Assistant + Lane Departure

Warning & Collision Warning Active Protection

Ford Active City Stop Forward Collision Warning with Brake Support

Mercedes-Benz

Pre-Safe Brake Collision Prevention Assist Distronic Plus

Peugeot LDWS - Lane Departure Warning System Toyota Pre-Crash System Volvo CitySafety

CWAB-PD

II. REVIEW OF EXISTING AEB SYSTEMS

CWAB-PD system, an active safety system,

was developed by Volvo Car Corporation. It has features of applying full brake in emergency scenario, providing warning and brake support to avoid pedestrian accidents [5]. The system’s third generation can help to decelerate at rate of 10 m/s2 to avoid collision at speed of 35 kph [5]. Similarly, Ford Active City Stop (ACS), a low-speed collision avoidance system, is able to prevent collisions at speed of up to 16 kph and reduce severity level at speed up to 32 kph [12]. The design of the system is based on the “Circle of Life” concept [5]. This means that statistical data of real-world accidents have been collected and analyzed to identify most common types of collisions, how often each collision happens and its causes before designing the system [5].

Targeting low-speed collision avoidance and

impact mitigation, Ford introduces ACS system with price from £250. Only 15 images per second are processed to identify potential hazards on the road. More advance, Audi is able to process up to 25 images per second in its latest AEB system installed on Audi A6 [13]. Light detecting and ranging sensor is used in the ACS system with setting scanning rate of 50 times per second. The system detects and calculates potential risk of hitting a stationary, slow-moving object. In case of emergency, it firstly pre-charges brake and wait for response from driver, if no action is taken, system applies brakes and reduce engine torque, followed by light up rear hazard lights [12].

Audi is leading in developing AEB systems

with various advanced technologies. Pre-Sense Front Plus is designed to avoid or mitigate rear-end accidents and can be applied to either moving or stationary objects. The system uses two long range radars and one windscreen-mounted camera for detection of potential


Proceedings of the


Page 40

SAEM 2013-010

collision with preceding objects. Front view of host vehicle is captured by camera at speed of 25 fps [13]. The system works at maximum speed of 200 kph, additionally warning signals are capable of working at speed even higher than 200 kph [14]. Adaptive Cruise Control (ACC) system is the most important and complex driver assistance system. It uses two long-range high-end radars with maximum operation range of 250 m and one interior-mirror-based camera with maximum operation range of 60 m. Hence control system is able to regulate host vehicle speed up to 250 kph. The system works at different traffic scenario such as changing lanes, passing or turning off at curvy roads, driving in city area. Pre-Sense Front and ACC system helps driver to prevent rear-end collision. Partial braking has effect to decelerate vehicle at rate of 3 m/s2 and further increases to deceleration rate of 5 m/s2 before full brake force is applied. Compared with 10 m/s2 deceleration capability of CWAB-PD system developed by Volvo, ACC system is not yet advanced [5].

Audi Pre-Sense Rear is the first AEB system

using two radars (max 70 m) to detect potential collision caused by back-end approaching vehicle. Night Vision Assistant system highlights detected pedestrians and animals using thermal imaging camera as far infrared system (FIR). The system helps to assist driver during night time. Volvo and Ford have offered pedestrian detection feature to their AEB systems but not yet developed AEB systems working at night [5, 12]. The camera operation range is up to 300 m however effective working range at the moment is 100 m. The camera is heated during cold weather condition and it is cleaned if dirty [15].

Except Ford, most of AEB systems are expensive and costly due to the use of radar sensor. Currently there is limitation in improvement of road safety in EU market for AEB systems due to cost and performance of sensor technologies [3].

One important feature that an active AEB

system needs to have according to Japanese Ministry of Land, Infrastructure and Transport [3] (MLIT) in Japan is that the “stopping distance” to avoid collision must be at least one meter [5]. Third generation CWAB-PD system

could not be able to achieve “stopping distance” of one meter [5].

Speed reduction is an important factor in

comparing impact mitigation performance of AEB systems. 10% reduction of speed before collision can reduce 30% fatal injuries; and in pedestrian collision, a reduction of 25 kph from 50 kph can decrease 85% risk for fatal injuries [5]. Audi ACC system could reduce speed of vehicle up to 40 kph when impact occurs [15] and Volvo CWAB-PD can reduce 10% of pre-crash speed [5]. Table 2 shows speed reduction of Audi and Volvo’s AEB systems.

Table 2: Speed reduction of some AEB systems

Automaker Speed Reduction

Audi Adaptive Cruise Control

40 kph (Max)

Volvo CWAB-PD

10 % of pre-crash speed

Common risk matric is time to collision or

time to impact [8, 15]. The calculation of the risk matric is very dependent on collision scenario and “unobservable factors”: (i) driver’ state and intention, (ii) surrounding objects and (iii) road condition [8]. In the combined Pre-Sense Front + ACC system, time to collision, an important factor in designing AEB system, is 0.5 s while in Mercedes-Benz and Volvo AEB systems, the time to collisions are 0.6 s and 1 s respectively [3, 5, 15]. Table 3 shows several values of time to collision used in AEB systems developed by Audi, Mercedes-Benz and Volvo. Table 3: Time to collision used in some AEB systems

Automaker Time to Collision (s)

Audi Pre-Sense Plus Adaptive Cruise Control

0.5

Mercedes-Benz Pre-Safe Brake Brake Assist Plus

0.6

Volvo CWAB-PD

1

II. AEB SYSTEMS AND FALSE DETECTION

ISSUES None AEB system is perfect; hence there is

certain error in detection of obstacles and false emergency alarm generated by the system. Unnecessary emergency braking maneuver would create “unacceptable false activation” to


Proceedings of the


1

SAEM 2013-010

driver [8]. Therefore, driver might feel less convenient while driving car and might deactivate the safety system to avoid the inconvenience. In this situation, AEB system has no effect on improving road safety.

Measurement noises, bad weather leading to

uncertainties in vehicle’s condition estimation, not considering road borders are factors that could result in ineffective AEB system [3, 8]. To avoid false alarm, emergency brake could only be triggered when driver already acknowledges the situation and applies force to brake. The acknowledgement might be sensed via change in brake’s pressure and gradient. The combination of both steering and braking maneuvers results in more efficient collision avoidance than just steering only and therefore less false alarms is generated [8]. However disadvantage of this combination is at the delay of system’s response due to the extension of collision avoidance cases in consideration and analysis. It is important to have a “multiclass classification” of objects such as animals, pedestrians, bikes, different types of vehicles to have a more efficient AEB system [5, 8, 15].

IV. FUTURE TRENDS OF AEB SYSTEMS

V2V and V2I communications allowing

wireless transmission of information between two vehicles on the road or between one vehicle and roadside infrastructure could help to improve road safety [3] by preventing 76% of crashes on roadway [16]. Sending wireless information among vehicles helps the vehicles to obtain travelling information of each other in 360o rather than just limited to line-of-sight as when using expensive radar or vision system [3, 16]. Travelling information such as location, position, velocity and acceleration from one vehicle can be sent to another vehicle on road [16, 17]. The information is processed by both cars and potential dangers could be detected. Then warning signals can be generated to catch attention of drivers to avoid accident. When accident occurs, vehicles involved in the accident get stuck on road and could be obstacles for other moving vehicles. It is dangerous if other drivers do not acknowledge that accident vehicles are in front of their trajectories. Hence it is expected that with V2V and V2I technologies, modern car could send accident information to approaching vehicles to alert their drivers about potential danger.

Additionally, modern vehicle could also send the information to infrastructures near roadside to notify about the accident, and provide necessary information for rescue service. Current V2V communication relies on radio communication at different frequency bands and faces signal distortion by “radio echoes” effect from objects existing on road side [18]. It is estimated that dedicated short range communication (DSRC) using IEEE 802.11p automotive W-Fi standard will be featured in 61.8 % new vehicles by 2027 [19]. However, the reliability of the communication standard in dense traffic scenarios such as in urban environment or traffic congestion is still questioned by researchers [17]. The communication issues are open areas for further research and development of V2V communication. Overall, future AEB system should offer V2V and V2I communications as basic needs for better accident avoidance and faster rescue service.

V. MALAYSIA’S TREND FOR

DEVELOPMENT OF AEB SYSTEMS Currently AEB systems are not developed in

Malaysia yet. No AEB system is offered for all products of two dominant automakers in Malaysia namely Proton and Perodua. Hence they should pay attention to the AEB research area in these coming years. In the design of AEB system, several factors should be placed in consideration.

A. Sensors’ Consideration

Laser range finder, camera, accelerometer

and speedometer are commonly used to obtain vehicle and surrounding’s information for AEB system to function. The first two sensors could be used in combination or separation to identify obstacle on road while the third and fourth ones are used to measure current acceleration and velocity of vehicle. Laser range finder has three ranges: short range (less than 1 meter), mid-range (less than 100 meters) and long range (over 100 meters) [4, 20-22]. Short range distance sensor is normally used for parking maneuvers while mid-range and long range distance sensors are used in AEB system [4]. Mid-range and long range laser range finder are excessively expensive (RM 20,622 for Model UTM-30LX, detection distance up to 30 m) and is not suitable to use in AEB system for


Proceedings of the


Page 42

SAEM 2013-010

Malaysian cars (Proton Exora’s unit price: RM 59,548 or Perodua Viva 660 BX’s unit price: RM 24,924) [23-27].

Since laser range finder is costly and not

suitable to be used in Malaysian cars, an alternative sensor would be camera. With the advancement of technologies, currently high quality cameras have affordable prices such as ABUS Security-Center TVIP51500 at RM 2,164 and resolution of 1280x1024 pixels [28]. There are two options for using camera as sensor to detect object in front of a car: day camera with daytime functioning capability and day/night camera with both daytime and nighttime functioning capability [29, 30]. There are several differences between CCTV camera and normal camera (e.g. DSLR, compact cameras). Normal camera can be used to take picture and then edit or process it after picture is taken (on and off operations) while CCTV camera mainly be used for safety monitoring purpose and offers real-time monitoring or operation per preset schedule (continuous operation) [31]. Therefore an affordable CCTV camera would be used as an alternative to detect obstacles on road.

Acceleration of an object indicates a rate of

change of its speed over a unit of time; acceleration’s SI unit is m/s2. In an AEB system, accelerometer measures acceleration of host vehicle. At the moment, there are about 90 choices for accelerometers offered by RS Malaysia Sdn. Bhd. with various unit prices from RM 6.55 for 3-axis digital accelerometer in model MAG3110FCR1 made by Freescale to RM 7,820.00 for 6-degree-of-freedom low-noise inertial sensor in model ADIS16375AMLZ manufactured by Analog Devices [28]. Selection of accelerometer for specific application such as in AEB system is suggested to be based on prioritized factors: (i) range of measurement in term of g (1 g ≈ 9.81 m/s2), (ii) measurement sensitivity (mV/g), (iii) response frequency (Hz), (iv) sensitive axis (single axis, tri-axis), and (v) size and weight of sensor [32]. Typical deceleration rate of AEB systems such as the first and second generations of Volvo CWAB-PD system is 5 m/s2 or about 0.5 g [5, 8]. Third generation of Volvo CWAB-PD system even could provide double the typical deceleration rate at 10 m/s2 or about 1 g [5]. With this information, accelerometer with measurement limit of 1.5 g (about 50% over maximum

deceleration rate in [5]) would be suitable for use in AEB system.

For AEB system, it is required to have a speedometer to obtain speed of vehicle. Speedometer can be categorized into two types: analog and digital. Analog speedometer is used more widely than digital one. However, AEB system needs digital speedometer in order to obtain vehicle speed in digital data form then send to control unit for processing. Digital automotive speedometers with two display options in kph and mph having unit price ranges from about RM 480 to RM 870 [33]. This range of price is considerably suitable for installation of digital speedometers on Malaysian cars.

B. Consideration of Important Parameters in Developing AEB Systems

Time to collision is the most common factor

for developing AEB systems worldwide. It is expected factor that all AEB systems should be able to determine when calculating potential of collision. Typical value of time to collision is less than 1 s. Stopping distance is essential for avoiding collision between host vehicle and target objects, it is suggested to have 1 m of stopping distance to avoid the collision. Additionally speed reduction is also needed when developing AEB systems targeting mitigation collision impact. Latest systems available in the market could achieve speed reduction of 40 kph when collision occurs. It is mentioned that a 10% reduction of speed before collision can reduce 30% fatal injuries; and in pedestrian collision, a reduction of 25 kph from 50 kph can decrease 85% risk for fatal injuries [5]. Therefore, developed system should be able to reduce speed of host vehicle by at least 10% when accident occurs.

C. Consideration of Development of Multiclass Object Detection System

Current technologies can identify potential

collision between (i) vehicle and vehicle, e.g. car with car, car with truck, car with motorbike, and (ii) vehicle and pedestrian e.g. car with pedestrian. Since objects involving in road traffic are (i) vehicles e.g. car (truck, van), motorbike, bicycle; (ii) people; (iii) animals for countryside area; it is important for AEB systems to avoid not only collision of host


Proceedings of the


Page 43

SAEM 2013-010

vehicle to target vehicle but also other objects presenting on the road.

VI. CONCLUSION

Overall, the paper presents intensive review

of existing AEB systems, its current development status and future trend. It is suggested that AEB systems to be developed in Malaysia should focus on using camera as sensor for distance measuring and avoid using laser range finder due to its excessive cost compared with local prices of current products offered by two major auto manufacturers in Malaysia namely Proton and Perodua. Multi-object detection algorithm should also be developed to enhance object detectability of AEB system. Weather conditions should also be taken into consideration because they are also the major cause of car accidents.

REFERENCES

[1] MIROS. (2013). General Road Accident Data in

Malaysia (1995 – 2010). Available: http://www.miros.gov.my/web/guest/road

[2] Z. M. Jawi, M. H. M. Isa, R. Sarani, W. S. Voon, and A. F. M. Sadullah, "Weather as a Road Safety Hazard in Malaysia - An Overview," 2009.

[3] F. J. Martinez, T. Chai-Keong, J. C. Cano, C. T. Calafate, and P. Manzoni, "Emergency Services in Future Intelligent Transportation Systems Based on Vehicular Communication Networks," Intelligent Transportation Systems Magazine, IEEE, vol. 2, pp. 6-20, 2010.

[4] A. Koelpin, G. Vinci, B. Laemmle, S. Lindner, F. Barbon, and R. Weigel, "Six-Port Technology for Traffic Safety," Microwave Magazine, IEEE, vol. 13, pp. 118-127, 2012.

[5] E. Coelingh, A. Eidehall, and M. Bengtsson, "Collision Warning with Full Auto Brake and Pedestrian Detection - a practical example of Automatic Emergency Braking," presented at the Annual Conference on Intelligent Transportation Systems, Funchal, 2010.

[6] Thatcham. (2013). Stop the crash. Available: http://www.thatcham.org/files/pdf/stop_the_crash_AEB.pdf

[7] Thatcham. (2011). AEB test procedures. Available: http://www.thatcham.org/safety/pdfs/AEB_test_procedures_nov11.pdf

[8] N. Kaempchen, B. Schiele, and K. Dietmayer, "Situation Assessment of an Autonomous Emergency Brake for Arbitrary Vehicle-to-Vehicle Collision Scenarios," IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, vol. 10, pp. 678 - 687, December 2009.

[9] A. Group. (2011). AEB Test Procedures - Autonomous Emergency Breaking. Available: http://www.thatcham.org/safety/pdfs/AEB_test_procedures_nov11.pdf

[10] E. NCAP. (2012). AEB Fitment Survey 2012. Available: http://www.euroncap.com/results/aeb/survey.aspx

[11] Peugeot. (2013). Driving Control. Available: http://www.peugeot.com/en/technology/safety/driving-control

[12] E. Ford Motor Company. (2012). New Ford Fiesta ‘Active City Stop’ safety tech captures and processes 15 images in the blink of an eye. Available: http://www.ford.co.uk/FordFleet/NewsAndReviews/FordForBusiness/2012/November/FiestaActiveCityStop

[13] Audi. (2013). Audi pre sense Anticipatory safety. Available: http://www.audi.com.au/au/brand/en/models/a6/a6_sedan/equipment/safety/audi_pre_sense.html

[14] E. Ncap. (2013). Reward 2012 - Audi Pre Sense Front Plus. Available: http://www.euroncap.com/rewards/audi_pre_sense_front_plus.aspx

[15] Audi. (2011). Driver assistance systems. Available: http://www.audiusanews.com/newsrelease.do;jsessionid=708C7D93E7D9BDFF95056C98D268E7C4?&id=2194&allImage=1&teaser=driver-assistance-systems&mid=18

[16] M. Schagrin. (2013). Vehicle-to-Vehicle (V2V) Communications for Safety. Available: http://www.its.dot.gov/research/v2v.htm

[17] Ü. Özguner. (2013). Vehicle-To-Vehicle Communication For Intelligent Transportation. Available: http://car.osu.edu/vehicle-vehicle-communication-intelligent-transportation

[18] G. C. Congress. (2013). New roadside scattering model to improve vehicle-to-vehicle communication for intelligent transportation systems. Available: http://www.greencarcongress.com/2013/07/v2v.html

[19] J.-P. Joosting. (2013). Vehicle-to-vehicle communications to reach 62% by 2027. Available: http://www.eetimes.com/document.asp?doc_id=1266622

[20] SICK. (2013). Long range distance sensors Available: http://www.sick.com/us/en-us/home/products/product_portfolio/distance_sensors/Pages/distance_sensors_long_range.aspx

[21] SICK. (2013). Mid range distance sensors. Available: http://www.sick.com/group/EN/home/products/product_portfolio/distance_sensors/Pages/distance_sensors_mid_range.aspx

[22] SICK. (2013). Short range distance sensors (displacement). Available: http://www.sick.com/group/EN/home/products/product_portfolio/distance_sensors/Pages/distance_sensors_short_range.aspx

[23] P. E. S. Bhd. (2012, June 28). Proton Exora. Available: http://www.proton-edar.com.my/Experience/Models/Exora/Overview.aspx

[24] Perodua. (2013, June 28). Alza Price. Available: http://www.perodua.com.my/ourcars/alza/price/gha

[25] Perodua. (2013, June 18). Myvi Price. Available: http://www.perodua.com.my/ourcars/myvi/price

[26] Perodua. (2013, June 28). Viva Price. Available: http://www.perodua.com.my/ourcars/viva/price

[27] C. T. S. Bhd. (2013, June 15). Laser Range Finder. Available: http://cytron.com.my/listProductGroup.php?pid=PR


Proceedings of the


Page 44

SAEM 2013-010

Y9EysWHScXLxwLEDcQGR0A!!!!!wTagIZ04QmaqPnXtZI=

[28] R. C. S. Bhd. (2013). Available: http://malaysia.rs-online.com

[29] BOSCH. Selecting the Right CCTV Camera. Available: http://www.boschsecurity.us/NR/rdonlyres/1A4F9B44-0376-4FC8-A735-151F02021082/0/SelectingtheRightCCTVCamera.pdf

[30] C. Ltd. (2013). CCTV camera buyers guide. Available: http://www.cctv42.co.uk/cctv-camera-buying-guide.aspx

[31] BORSCHE. (2012). Difference between CCTV Camera and Normal Camera. Available: http://www.borschecctv.com/en/news/Difference-between-CCTV-Camera-and-Normal-Camera-185.html#.UcQkwPnVDzI

[32] SENSR. (2013), Practical guide for accelerometers. Available: http://www.sensr.com/pdf/practical-guide-to-accelerometers.pdf

[33] I. Dakota Digital. (2013, June). Speedometers / Tachometers. Available: http://www.dakotadigital.com/index.cfm/page/ptype=results/Category_ID=240/home_id=59/mode=cat/cat240.htm

a review of automated emergency braking system and the ... … · “automatic emergency braking...

Documents