Nof S.Y. Springer Handbook of Automation

Подождите немного. Документ загружается.

1165

Vehicle and R

66. Vehicle and Road Automation

Yuko J. Nakanishi

Presently in the USA, Europe, Japan, and in other

parts of the world, intelligent transportation sys-

tem (ITS) technologies are being developed and

deployed to increase the intelligence of vehicles.

The two key beneﬁts of these technologies are en-

hancement of safety and mobility of the traveling

public. The intelligence is provided by electronics,

communications systems, software, and human–

machine interfaces and is assisting drivers with

many aspects of the driving task. Drivers may be

warned about potential crashes with other cars,

about objects that are hidden from their vantage

point, and about excessive speeds. Information

about real-time trafﬁc conditions including inci-

dents on a driver’s preferred route, travel times to

speciﬁc destinations, and about restaurants, ho-

tels, and other destination points may be provided.

In-vehicle navigation systems can tell drivers how

to get to a destination on a turn-by-turn basis

and may be linked to a central dispatch cen-

ter to summon help automatically in case of an

accident.

The major initiatives and technologies being

developed in the USA – integrated vehicle-based

safety systems, forward collision warning systems,

road departure crash warning systems, vehicle

infrastructure integration – are described and

discussed in this chapter. In addition, how they

66.1 Background .........................................1165

66.1.1 USA – Intelligent Transportation

Systems (ITS) Background ............ 1165

66.1.2 European Union –

Telematics Initiatives .................. 1169

66.1.3 Japan – ITS Initiatives................. 1170

66.2 Integrated Vehicle-Based Safety Systems

(IVBSS) .................................................1171

66.2.1 IVBSS Systems Architecture........... 1171

66.2.2 Forward Collision Warning (FCW)

System...................................... 1173

66.2.3 Road Departure Crash Warning

(RDCW) System ...........................1174

66.2.4 Human Factors........................... 1174

66.3 Vehicle Infrastructure Integration (VII)....1176

66.3.1 VII Beneﬁts................................ 1176

66.3.2 VII Risks .................................... 1177

66.4 Conclusion and Emerging Trends ...........1177

66.5 Further Reading ...................................1178

References ..................................................1180

interact with the driver is important in terms of

safety, liability, and acceptance of the technolo-

gies. The human factors elements that should be

considered are presented and discussed in the

chapter as well.

66.1 Background

66.1.1 USA – Intelligent Transportation

Systems (ITS) Background

In the USA, there were 5973000 police-reported motor

vehicle trafﬁc accidents in 2006. These crashes resulted

in 42642 deaths and 2575000 injuries. During 2006,

an average of 117 persons died daily as a result of

these crashes. There were another 4 189000 accidents

involving property damage only [66.1]. While the fatal-

ity rate has decreased from 1.69 per 100million vehicle

miles of travel (VMT) in 1996 to 1.41 in 2006, these

statistics are still staggering and reﬂect the very serious

economic and human costs of these accidents. (Na-

tional HighwayTrafﬁc Safety Administration (NHTSA,

Part G 66

1166 Part G Infrastructure and Service Automation

http://www.nhtsa.dot.gov/) reports that, in 2000, the

economic costs of trafﬁc accidents were US$230.6 bil-

lion.)

The latest mobility research by the Texas Trans-

portation Institute concluded that trafﬁc conges-

tion continues to worsen in US cities and causes

4.2billion hours of delays, depletes US$78 billion from

the US economy, and wastes 2.9billion gallons of fuel.

In 2005, congestion in urban areas caused the average

peak period traveler to spend an extra 38h of travel time

and consume an additional 26 gallons of fuel, at a cost

of US$710 per traveler [66.2].

Intelligent transportation systems (ITS)istheap-

plication of electronics, communications, information,

and information technology to transportation systems.

The primary drivers of these technologies are safety

and mobility. Concomitant beneﬁts are efﬁciency, con-

venience, and accessibility. Highlights of safety and

mobility beneﬁts generated by ITS projects are pre-

sented below [66.3].

Safety beneﬁts:

•

In Georgia, the Navigator incident management pro-

gram reduced secondary crashes from an expected

676 to 210 in the 12 months ending April 2004

(2006).

•

In North Carolina, a work zone equipped with smart

work zone traveler information systems observed

fewer crashes compared with other work zones

without the technology (2005).

•

In Baltimore, a second train coming warning sys-

tem decreased the frequency of the most common

risky behavior at crossings (i.e., drivers that crossed

the tracks after the protection gates began to as-

cend from the ﬁrst train before the protection gates

could be redeployed for the second train) by 26%

(2002).

•

Evaluation indicated that integrating dynamic mes-

sage signs (DMS) and incident management sys-

tems could reduce crashes by 2.8%, and that

integrating DMS and arterial trafﬁc control systems

could decrease crashes by 2%, in San Antonio, TX

(2000).

•

In Georgia, call boxes installed on a 39mile section

of I-185 were estimated to eliminate one injury per

year, and one fatality every 5years (2000).

•

An advanced curve warning system on an interstate

route in northern California caused over 68% of

drivers to reduce their speed (2000).

•

In a rural area of Virginia, a collision countermea-

sure system installed on a two-way stop-controlled

intersection reduced vehicle speeds by 2.4mi/h,

and increasedthe average projectedtime to collision

from 2.5to3.5s (2000).

•

An automated enforcement systems in California

decreased highway–rail grade crossingviolations by

up to 92% (1999).

•

A dynamic truck downhill speed warning system

installed on I-70 in Colorado reduced the average

speed of passing trucks by approximately 5.2mi/h

(1999).

•

In Colorado, a downhill truck speed warning sys-

tem installed on I-70 reduced runaway ramp usage

by 24% and contributed to a 13% drop in crashes

involving trucks and excessive speeds (1997).

•

After a ramp rollover warning system was installed

at three curved exit ramps on the beltway around

Washington, DC, there were no accidents at any

of these sites during the 3year postdeployment test

period evaluated (1997).

•

Following deployment of the TransGuide freeway

management system in San Antonio, TX, crash fre-

quency was reduced by 41% and incident response

time decreased by 20% (1997).

•

Advanced trafﬁc management systems in Amster-

dam and Germany reduced crash rates by 20–23%

(1999).

•

In Japan, a real-time incident detection and warning

system installed on a dangerous curve on the Han-

shin Expressway decreased the rate of secondary

crashes by 50% (1997).

Mobility beneﬁts [66.3]:

•

In Georgia, the Navigator incident management pro-

gram reduced the average incident duration from

67minto21min,saving7.25million vehicle-hours

of delay over one year (2006).

•

In Georgia, the highway emergency response

operator (HERO) motorist assistance patrol pro-

gram and Navigator incident management activities

saved more than 187million USdollars, yielding

a beneﬁt-to-cost ratio of 4.4:1 (2006).

•

In Utah, incident management teams in Salt Lake

Valley area decreased incident duration by approxi-

mately 20 min perincident on threemajor interstates

(2004).

•

In 2002, the Maryland coordinated highways

action response team (CHART) highway inci-

dent management program reduced delay by

about 30 millionvehicle−hours and saved about

5million gallons of fuel (2003).

Part G 66.1

Vehicle and Road Automation 66.1 Background 1167

•

In Albuquerque, NM, work zone surveillance and

response at the Big I Interchange reduced average

clearance time by 44% (2001).

•

During the ﬁrst year of operations at the Big I work

zone in Albuquerque, temporary trafﬁc management

and motorist assistance patrols reduced the average

incident response time to less than 8 min, and no

fatalities were reported (2001).

•

Modeling performed as part of an evaluation of nine

ITS implementation projects in San Antonio, TX,

indicated that integrating dynamic message signs

(DMS), incident management, and arterial trafﬁc

control systems could reduce delay by 5.9% (2000).

•

The delay reduction beneﬁts of improved incident

management in the Greater Houston area saved

motorists approximately US$8440 000 annually

(1997).

•

In San Antonio, TX, a freeway management

system led to an estimated delay savings of

700vehicle−hours per major incident (1997).

•

In Brooklyn, an incident management system on

the Gowanusand Prospect Expressways used closed

circuit television (CCTV), highway advisory radio,

DMS, and a construction information hotline to im-

prove average incident clearance time by about 1 h,

a 66% improvement (1997).

The National ITS Architecture [66.4] provides a com-

mon framework for the planning deﬁning and imple-

mentation of ITS, and deﬁnesthe requiredfunctions, the

physical entities or subsystems where these functions

reside, and the information ﬂows and data ﬂows link-

ing these functions and physical subsystems together.

The National ITS Architecture consists of three layers,

institutional, communications, and transportation:

•

Institutional layer: represents the existing and

emerging institutional constraints and arrangements

and addresses policy issues, funding incentives,

working arrangements, and jurisdictional structure.

•

Communications layer: comprises communication

equipment such as wireless transmitters and re-

ceivers.

a) b) c)

Fig. 66.1 (a) Collision avoidance system. (b) Driver assis-

tance system.

•

Transportation layer: presents relationships among

the transportation-related elements, including trav-

eler subsystems, vehicles, transportation manage-

ment centers ﬁeld devices, and external system

interfaces.

Logical architecture deﬁnes ITS processes, process-

to-process data ﬂows, and data elements and is not

technology speciﬁc. Physical architecture is the high-

level structure containing major ITS system compo-

nents. Subsystems are the structural elements of the

physical architecture. The subsystems are categorized

into center, traveler, ﬁeld, and vehicles [66.5].

The intelligent vehicles subsystem is comprised of

collision avoidance systems, driver assistance systems,

and collision notiﬁcation systems (Fig.66.1).

Collision avoidance systems (CAS) warn drivers of

impending danger. The vehicle warnings need to occur

early enough for the driver to take action, and must also

be clear and understandable to the driver. Since multi-

ple alarms would be confusing and distracting for the

driver, a management system needs to be present to de-

cide which CAS or driver assistance warning to provide

when the condition warrants it. CAS systems comprise

the following elements:

•

Intersection collision warning systems detect and

warn drivers of approaching trafﬁc at high-speed

intersections; these systems use both vehicle- and

infrastructure-based technologies. Once vehicle in-

frastructure integration is in place, other drivers

would also be able to receive warnings about im-

pending violations (Fig.66.2).

•

Obstacle detection systems use vehicle-mounted

sensors to detect obstructions, such as other vehi-

cles, road debris or animals, in a vehicle’s path and

alert thedriver. Anexample of a more advanced sys-

tem for obstacle detection and road navigation in

intelligent vehicles is presented in Fig.66.3.

•

Lane-change warning systems have been deployed

to alert bus and truck drivers of vehicles or obstruc-

tions in adjacent lanes when the driver prepares to

change lanes.

•

Lane departure warning systems warn drivers that

their vehicle is unintentionally drifting out of the

lane.

•

Rollover warning systems notify drivers of heavy

trucks when they are traveling too fast for an

approaching curve, given their vehicles operating

characteristics. This system would also be useful for

light trucks (sports utility vehicles (SUVs)) as well.

Part G 66.1

1168 Part G Infrastructure and Service Automation

a) b) c)

d) e) f)

Fig. 66.2 (a) The robot approaches an intersection and identiﬁes a waiting car (rectangle marked by a blue diamond).

As the gets closer

(b) it detects a second vehicle (rectangle, left of frame), and upon coming to a stop (c) tracks a third

vehicle, which arrives late (rectangle middle of frame).

(d) The ﬁrst vehicle begins moving through the intersection, and

precedence is transferred to the second vehicle, denoted by the teal diamond.

(e) The second vehicle begins traversing the

intersection.

(f) Upon all clear, the robot takes precedence ahead of the third vehicle and safely navigates the intersection

(courtesy of Tartan Racing, Carnegie Mellon University)

•

Road departure warning systems have been tested

using machine vision and other in-vehicle systems

to detect and alert drivers of potentially unsafe lane-

keeping practices and to keep drowsy drivers from

running off the road.

•

Forward collision warning systems use microwave

radar and machine vision technology to help de-

a) b)

c) d)

tect and avert crashes. Drivers are warned of unsafe

conditions via vehicle displays or audible alerts. If

a driver does not properly apply brakes in a crit-

ical situation, some systems automatically assume

control and apply the brakes in an attempt to avoid

a collision.

•

Rear-impact warning systems use radar detection to

prevent accidents. A warning sign is activated on

the rear of the vehicle to warn tailgating drivers of

impending danger.

Driver assistance systems (DAS) assist drivers nav-

igate their vehicles safely, enhance their vision, and

control the speed of their vehicles. These systems will

Fig. 66.3 (a) Vehicle navigating down a single-lane road

(lane boundaries shown in blue, current curvature of the

vehicle shown in pink, minimum turning radius arcs shown

in yellow).

(b) extracted centerline of the lane (in red).

its current state, the centerline path, and lane bound-

aries. A single trajectory is selected for execution, as

discussed.

(d) the same scenario from an alternative view-

ing angle (courtesy of Tartan Racing, Carnegie Mellon

University)

Part G 66.1

Vehicle and Road Automation 66.1 Background 1169

be especially valuable in inclement weather or in other

hazardous conditions. Driver assistance systems com-

prise:

•

In-vehicle navigation and route guidance systems

with global positioning system (GPS) technology

may reduce driver error, increase safety, and save

time by improving driver decision in unfamiliar ar-

eas.

•

Integrated driver communication systems enable

drivers and dispatchers to coordinate rerouting deci-

sions on-the-ﬂy and can also save time and money,

and improve productivity.

•

In-vehicle vision enhancement improves visibility

for driving conditions involving reduced sight dis-

tance due to night driving, inadequate lighting, fog,

drifting snow or other inclement weather condi-

tions.

•

Object detection systems warn the driver of an ob-

ject (front, side or back) that is in the path or

adjacent to the path of the vehicle. Currently the

most common application is parking aids for pas-

senger vehicles. Toyota’s intelligent parking assist

(IPA) system uses cameras and sensors to steer the

car for the driver for parallel parking or backing into

a parking space. Obstacle detection sensors for rear

and forward bumpers are available from automakers

or available from individual vendors.

•

Adaptive cruise control systems maintain a driver

set speed without a lead vehicle, or a speciﬁed

following time if there is a lead vehicle and it is

traveling slower than the set speed.

•

Intelligent speed control systems limit maximum

vehicle speed via a signal from the infrastructure to

an equipped vehicle.

•

Lane keeping assistance systems can make minor

steering corrections if the vehicle detects an immi-

nent lane departure without the use of a turn signal.

•

Roll stability control systems take corrective action,

such as throttle control or braking, when sensors de-

tect that a vehicle is in a potential rollover situation.

•

Drowsy driver warning alerts drivers when they

have become fatigued.

•

Precision docking systems automate precise posi-

tioning of vehicles at loading/unloading areas.

•

Coupling/decoupling: intelligent cruise control,

speed control, guidance/steering, and coupling/de-

coupling systems which help transit operators link

multiple buses or train cars into trains each as-

sist drivers with routine tasks that weight on driver

workload.

•

Onboard monitoring tracks and reports cargo con-

dition, safety, and security, and the mechanical

condition of vehicles equipped with in-vehicle di-

agnostics. This information can be presented to the

driver immediately, transmitted off-board, or stored.

In case ofa crashor near-crash, in-vehicle eventdata

recorders can record vehicle performance data and

other input from cameras or sensors.

Collision notiﬁcation systems (CNS)suchasGM’s

Onstar system report accidents to emergency respon-

ders or dispatch centers either manually (MAYDAY)

or automatically with automatic collision notiﬁcation

(ACN). Details about the accident such as its severity

and number of passengers inside the vehicle may also

be reported.

•

MAYDAY/ACN products utilize location technology,

wireless communication, and a third-party response

center to notify the closest public safety answering

point (PSAP) for emergency response.

•

Advanced ACN systems use in-vehicle crash sen-

sors, GPStechnology, and wireless communications

systems to supply public/private call centers with

crash location information, and in some cases, the

number of injured passengers and the nature of their

injuries.

Additional information and ITS initiatives in the

USA are providedin subsequentsections ofthis chapter.

66.1.2 European Union –

Telematics Initiatives

In Europe, telematics is the equivalent of ITS technolo-

gies, and the key beneﬁts being sought by the European

countries are similar. The European Commission seeks

to further reduce road fatalities by 50% by 2010. To

accomplish this, preventive safety systems are being de-

veloped within various Europe-wide initiatives such as

eSafety and PReVENT.

The eSafety effort established by the European

Commission contains nine working groups which focus

on crash-causation analysis, human–machine interface,

and others. First-generation or autonomous systems

within the vehicle act as isolated units with minimal in-

teraction among the units and minimal external input.

The core technologies are:

•

Advanced vehicle sensors and communications sys-

tems that integrate with onboard navigation systems

•

Advanced GPS coordinates with digital mapping

and location-based services

Part G 66.1

1170 Part G Infrastructure and Service Automation

•

Vehicle and infrastructure information,emphasizing

safety and efﬁciency

Second-generation initiatives such as PReVENT

manage the vehicle’s performance in the context

of the overall system, with the vehicle and infras-

tructure working together to enhance the safety of

motorists [66.6].

PReVENT systems are being designed to be in-

teroperable across European nations and even on

a global basis. PReVENT technologies will predict

the behavior of trafﬁc and other drivers, and include

sensing technologies, in-vehicle digital map and po-

sitioning technologies, and wireless communications

systems. The technologies revolve around provision

of driver-oriented safety applications. A key element

of PReVENT is advanced driver assistance systems

(ADAS), whichwas originally initiated as aresult of the

EUREKA programme for an European trafﬁc of highest

efﬁciency and unprecedented safety (PROMETHEUS)

project [66.7].

•

The 3year MAPS&ADAS subproject has produced

an ADAS interface, ADAS maps for PReVENT test

sites, and a safety impact assessment of the maps.

This project isneeded becauseaccess to map databy

applications other than navigation requires a stan-

dard nonproprietary interface.

•

The ProFusion subproject focuses on sensors and

sensor data fusion. The ﬁrst preliminary concept

building phase has been completed. The objectives

of the second phase are:

– To develop concepts and methods for sensor

data fusion to improve the reliability and robust-

ness of perception systems

– To deﬁne and implement a ﬂexible and modu-

lar architecture for fusion systems with diverse

approaches

– To establish and maintain the Fusion Forum,

which promotes sensor data fusion in driver as-

sistance and active safety systems

•

RESPONSE 3 will develop a code of practice for

the development and testing of ADAS for European

industry and will address the key issues of reason-

able safety and duty of care by deﬁning safe ADAS

development and testing.

•

PReVAL subproject will provide PReVENT with

a common evaluation framework and methodology

to assess the impact of various applications.

•

INSAFES (integrated safety system) is developing

new control functions to improve the functionality

of applications already developed.

•

The 3 year WILLWARN (wireless local danger

warning) subproject will generate a safety applica-

tion that warns the driver whenever a safety-related

critical situation occurs beyond the driver’s ﬁeld

of view. The application will include the develop-

ment of onboard hazard detection, in-car warning

management, and decentralized warning distribu-

tion by vehicle-to-vehicle communication on a road

network

•

The SASPENCE (safe speed and distance) system

will guide the driver regarding desirable speed and

headway for the given driving conditions.

66.1.3 Japan – ITS Initiatives

In Japan, the focus of ITS efforts has initially been on

in-vehicle information systems, and the beneﬁts sought

have been in terms of safety and the environment. As in

Europe, there has been strong national leadership with

regards to achieving these beneﬁts. In comparison with

other nations, Japan has many vendors of ITS-related

products and services.

The vehicle information and communication system

(VICS) is a real-time system that provides information

including weather, road and trafﬁc conditions, and nav-

igation assistance to in-vehicle navigation systems. The

system has been considered a great success with more

than 9million subscribers in 2004. While the VICS ser-

vice is free, users pay a one-timesetup fee,and purchase

the onboard navigation system. This system serves as

a catalyst for further ITS deployment in automobiles.

Taxi probes equipped with GPS and wiper sensors are

used by the Japan Road Trafﬁc Information Center to

obtain information about the transportation system; the

information is then analyzed and passed along to mo-

torists using VICS. A mobile radio local-area network

in the 2.4GHz bandwidth was installed by the Japan

Highway Public Corporation and is used to commu-

nicate both data and images to in-vehicle navigation

systems [66.6].

Other technologies that have been developed in-

clude the following:

•

Vehicles equipped with cameras to allow drivers

to see objects and roadway features in blind spots.

This technology is an element of advanced cruise-

assisted highway systems (AHS), which provide

information on highway features (curve data or

other roadway information) as well as warnings of

obstacles. AHS warns drivers of impending dangers

and allows a timely response to them.

Part G 66.1

Vehicle and Road Automation 66.2 Integrated Vehicle-Based Safety Systems (IVBSS) 1171

•

The radar cruise control (with low-speed following

mode) system enables drivers to adjust headway be-

tween their own vehicle and the one immediately

ahead, thereby providing uniform trafﬁc ﬂow and

automatic adjustment of speed based on the actions

of the lead vehicle. Adaptive cruise control provides

the added beneﬁt of mild to increasingly intense

braking if the driver of a vehicle does not respond

safely to a slowing or stopping vehicle ahead.

The remainder of the chapter will discuss speciﬁc

ITS initiatives established in the USA.

66.2 Integrated Vehicle-Based Safety Systems (IVBSS)

The Integrated Vehicle-Based Safety Systems (IVBSS)

initiative addresses the subset of accidents caused by

rear-ending, road departures, and lane changes. The

initiative involves the development of objective tests

and criteria for performance of IVBSS, and how best

to communicate an integrated warning to the driver.

IVBSS is expected to have the potential to reduce these

accidents by almost 50%, which translates into about

21000 lives saved per year.

The IVBSS program will evaluate the safety ben-

eﬁts and driver acceptance of the following crash

warning subsystems [66.8]:

•

Forward crash warning, which warns light vehicle

and heavy truck drivers of the potential for a rear-

end crash with another vehicle

•

Lateral drift warning, which warns light vehicle and

heavy truck drivers that they may be drifting inad-

vertently from their lane or departing the roadway

•

Lane-change/merge warning, which warns light ve-

hicle and heavy truck drivers of possible unsafe

lateral maneuvers based on adjacent or approach-

ing vehicles in adjacent lanes, andincludes full-time

side object presence indicators

•

Curve-speed warning, which warns light vehicle

and heavy truck drivers they may be driving too

quickly into an upcoming curve

The two main functions of IVBSS are situation

characterization, which determines that a potential

crash threat exists, and threat assessment, which de-

termines whether an alert should be issued for each

warning subsystem. The major elements of IVBSS are

as follows [66.8].

•

Sensing subject vehicle information and driver con-

trol inputs: The IVBSS system must obtain certain

information from the vehicle; for example, for rear-

end crashes, vehicle speed, yaw rate, and driver

brake switch are important.

•

Sensing roadway geometry and characteristics:Ob-

taining information about the roadway (e.g., road

curvature, vehicle axes relative to the lane, position

of the vehicle in the lane, determination of whether

the lane edges are road edges, time rate of change of

the lateral position of the vehicle relative to the road

edge).

•

Sensing objects and characterizing object type and

motion: To avoidcrashes, identiﬁcation and location

of other vehicles in the same lane and in adjacent

lanes are important.

•

Estimating road condition parameters: Each warn-

ing function is required to obtain and use available

data that may indicate low road friction.

•

Sensing driver attributes: In the subsequent year of

the research, individual driver behavior will be in-

tegrated into decisions about the issuance of crash

alerts.

US Department of Transport (DOT)-sponsored ef-

forts have produced driver–vehicle interfaces including

visual and auditory display requirements; speciﬁc warn-

ing messages have been identiﬁed and are under

development; human factors tests were conducted to

test the interfaces and the system itself; and prototype

hardware was developed for the IVBSS evaluation.

US DOT-sponsored ﬁeld tests of forward crash

warning (FCW) systems for light vehicles in the Colli-

sion Avoidance Metrics Partnership (CAMP) program

which produced performance criteria for radar-based

FCW crash warning systems [66.9]. Further FCW

crash warning systems research was performed in the

Automotive Collision Avoidance Systems (ACAS)pro-

gram [66.10–12].

66.2.1 IVBSS Systems Architecture

The systems architecture diagram shown in Fig. 66.4

displays the major subsystems along with the sensors

and software, and the hardware interfaces and commu-

nication protocols of the IVBSS.

The ﬁve major elements of the architecture (to the

left of the buses) are:

Part G 66.2

1172 Part G Infrastructure and Service Automation

Headrest

speakers

Digital

map

Warning

enabled

Inertial

sensors

Driver

face

camera

Cellular

antenna

Shared

cameras

GPS

antenna

Rear

right

back

Rear

left

back

Short-range radar

Short-range

forward

camera

(CMOS)

Front long

range

camera

(CMOS)

IVBSS system CAN bus (600K)

LCM CAN bus (600K)

FCW CAN bus (600K)

Vehicle CAN bus – high speed

Vehicle CAN bus – low speed

Snooze Volume

Sensi-

tivity

CMOS

camera

Gateway

Vehicle

stability

assist

Right side vision module

CMOS

camera

Left side vision module

Lateral drift

warning

(LDW)

module

AMP

Side

mirror

(icons)

OEM

audio

system

Haptic

controller

Curve speed

warning

(CSW)

module

Lane change/merge

FCW

arbitration

DVI (LFAD) module

Data acquisition

system (DAS)

module

Forward

vision

platform

Bus legend

Range processing unit (RPU)

Front

left

side

Front

right

side

Rear

left

side

Front long

range radar

Rear

right

side

Vehicle CAN bus – low speed

Vehicle CAN bus – high speed

FCW CAN bus (600K)

LCM CAN bus (600K)

IVBSS system CAN bus (600K)

IVBSS

driver

display

Audio

controller

Yaw rate

sensor

Fig. 66.4 IVBSS system architecture (after [66.8])

•

Gateway: Translatesappropriate messagesfrom two

original equipment manufacturer (OEM) data buses

to one of the project CAN buses

•

Lateral drift warning module: Uses forward vision-

based lane tracking and other signals from the CAN

bus to broadcast LDW (lane departure warning)

alert requests onto the bus

•

Curve speed warning module:UsesGPS, an on-

board digital map, and other information to broad-

cast CSW alert requests onto a serial link to the

LFAD (light-vehicle module for LCM, FCW, arbi-

tration, and DVI) module

•

LCM (lane change/merge warning)/FCW/arbitra-

tion/DVI (driver-vehicle interface) module: A chas-

sis that includes processors and other hardware

on which LCM, FCW, arbitration, and DVI are

hosted

•

Data-acquisition system module: A two-central pro-

cessing unit (CPU) module with peripherals that

records data.

Part G 66.2

Vehicle and Road Automation 66.2 Integrated Vehicle-Based Safety Systems (IVBSS) 1173

The additional elements above the three IVBSS

CAN buses include two vision-based modules that as-

sist with LCM functionality on the left and right side

of the vehicle and a third vision-based module that as-

sists FCW target selection and three pairs of short-range

radars communicating with the IVBSS CAN buses

through a radar processing unit (RPU).

66.2.2 Forward Collision Warning (FCW)

System

Forward collision warning (FCW) systems warn the

driver when the vehicle is about to crash into the rear-

end of another vehicle or object. FCW systems use

sensors such as radar, GPS, and vehicle-to-vehicle com-

munications systems in combination with appropriate

algorithms to detect static and moving obstacles and

determine when to issue an alert request. The request

is forwarded to an arbitration subsystem which re-

ceives requests from all driver warning systems and

decides which warning to issue. The two important

driver behavior parameters are: driver brake reaction

time – the time required for the driver to respond to

the alert by braking – and driver deceleration behavior

in response to the alert under a wide range of condi-

tions [66.9].

Figure 66.5 depicts a cooperative FCW application.

The host vehicle, equipped with the FCW system,

can transmit and receive dedicated short range commu-

nications (DSRC) messages while only target vehicle 2

Fig. 66.5 Cooperative forward collision warning applica-

tion (www.nhtsa.gov)

has a similar DSRC messaging system. The host ve-

hicle’s system will scan vehicles in its sensing area,

choose the closest in-path target, and analyze the tar-

get’s dynamics information. If the target is a threat,

a warning will be sent to the driver of the target vehicle.

Currently, FCW updates at 100ms and have coverage

ranges of 150m.

The system performance is enhanced by vehicles

that have DSRC due to the additional redundancy pro-

vided in vehicle detection. Redundancy is important

especially in adverse environmental (e.g., snow) con-

ditions. Dynamics information received from the target

vehicle is richer; information about vehicle size, type,

position, and turn-signal status can be transmitted to the

host vehicle. Also, the information is received with low

latency. Earlier detection of the target vehicle cutting

into the host vehicle’s path is possible due to informa-

tion about turn-signal status, steering wheel position,

and yaw rate. Conversely, false alarms may be reduced

by earlier detection of the target vehicle leaving the host

vehicle’s path.

Another beneﬁt of DSRC-equipped vehicles is early

detection of stopped vehicles. Presently, it is not possi-

ble to identifystopped vehicles untilit is too lateto issue

a warning toprevent anaccident. A DSRC-equipped ve-

hicle can warn other vehicles that it is stopped. Once all

vehicles on the road are equipped with DSRC, sensors

would not be needed.

A Forward Collision Warning Requirements Project

sponsored by NHSTA sought to develop a FCW crash

alert timing approach by examining last-second brak-

ing and lane-change maneuvers. One ﬁnding was that

differences in last-second braking and steering were de-

pendent on kinematic conditions.

Two braking onset models were developed – one

using linear regression, and the other using logistic re-

gression.

1. The ﬁrst model develops a deceleration value above

which the driver is assumed to be in alert mode

when the driver is about to brake. The inputs of the

model werethe closing speed between the following

and leading vehicle, the lead vehicle deceleration

value, and knowledge of whether or not the lead

vehicle is moving or stationary.

2. The second model is the three-tiered inverse time-

to-collision model which predicts the probability

that a driver is in an appropriate alert condition. The

model contains inverse time to collision which is the

difference in speeds of the following and leading

vehicles divided by the ranges between the vehicles.

Part G 66.2

1174 Part G Infrastructure and Service Automation

The FCW software modules are as follows:

1. Radar-based scene tracking: Tracks objects with re-

spect to the subject vehicle

2. Path prediction and data fusion: Determines the up-

coming geometry

3. Primary target determination: Determines the in-

lane primary target thatis considered the most likely

to pose a crash threat

4. Vision-based primary target validation and charac-

terization: Validates the choice of primary target,

and includes veriﬁcation that the target is a relevant

vehicle

5. Threat assessment: Given the primary target, de-

cides whether to issue an FCW crash alert request

6. FCW false-alarm management: Manages false

alarms to reduce the nuisance alarm rate using his-

torical data and subsequent driver responses

66.2.3 Road Departure Crash Warning

(RDCW) System

Road departure accidents make up less than one-ﬁfth

of accidents but cause almost 40% of annual highway

fatalities and are often caused by driver inattention,

intoxication or drowsiness. Road departure crash warn-

ing (RDCW) systems warn drivers of impending road

departure crashes due to lateral drift (unintentionally

moving into another lane) or excessive curve speeds.

The Intelligent Vehicle Initiative (IVI)RoadDe-

parture Crash Warning System Field Operational Test

(RDCW FOT) project conducted under a cooperative

agreement between the USDepartment of Transporta-

tion and the University of Michigan Transportation

Research Institute and its partners provided valuable in-

sights into driver behavior and the speciﬁc beneﬁts of

the RDCW system which may be generalized to sim-

ilar systems. The RDCW system tested in the RDCW

FOT project targeted lane departure crashes as well as

vehicles traveling into curves at unsafe speeds [66.13].

The RDCW system architecture developed for the

project is shown in Fig.66.6.

The lateral drift warning system (LDW) provided

a set of driver alerts in the form of visual and haptic

seat vibrations when the vehicle started to drift towards

or outside of the lane boundaries; the driver was then

expected to assess the situations and take appropriate

action. The LDW system issued imminent alerts in the

form of a visual icon and audible message in more haz-

ardous conditions (e.g., when striking an object was

imminent).

The key primary sensors used by the LDW system

were forward cameras to identify the lane boundaries

and road edges, brake switch, and turn-signal switch.

Supporting equipment were GPS, a digital map and

database, vehicle speed, yaw-rate gyro, and forward-

and side-looking radars.

The three major changes effected by the LDW sys-

tem on driver behavior were all positive; these changes

were:

1. The rate of turn-signal use during lane changes in-

creased.

2. Lane keeping performance improved: The standard

deviation of lane position decreased signiﬁcantly:

the number of events in which the outside of the

vehicle’s tire crossed the lane edge or came within

4inches of the lane edge was reduced by 50%. The

time spent within 4 inch of the lane edge or outside

the lane edge was reduced by 63%.

3. With the warning system in place, the data suggest

that the vehicle returns to the lane more quickly.

The curve speed warning system (CSW) warns

drivers to reduce speed before entering an upcoming

curve. The CSW system uses GPS and a digital map to

determine curve locations and radii. Recent driver ac-

tions such as applying turn signals were incorporated

into the system’s decision to issue an alert. The driver

warnings consisted of visual, audible, and haptic cues

similar to those of the LDW system.

In addition to GPS and a digital map, the CSW

system used vehicle speed and yaw-rate gyro as the pri-

mary sensors, while thesupporting equipment consisted

of the camera, brake switch, and turn-signal switch.

66.2.4 Human Factors

Background

Human factors research is an important aspect of vehi-

cle automation systems. Evaluation of driver interaction

with driver assistance systems such as in-vehicle navi-

gation systems, evaluation of different user interfaces

for the systems, and driver understanding of and reac-

tion times to collision warning systems are essential

in the development of a safe and effective technology.

According to NHTSA, driver distraction contributes to

20–30% of crashes, and in-vehicle navigation systems

and other systemsthat requiredriver attention willcause

driver distraction. A 2000 NHTSA study conﬁrmed that

entering destinationsinto navigation systems along with

cellphone dialing and radio tuning were distracting to

the driver [66.14].

Part G 66.2