[Federal Register Volume 59, Number 148 (Wednesday, August 3, 1994)]
[Unknown Section]
[Page 0]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 94-18802]


[[Page Unknown]]

[Federal Register: August 3, 1994]


-----------------------------------------------------------------------

DEPARTMENT OF TRANSPORTATION
National Highway Traffic Safety Administration

49 CFR Part 571

 

Lamps, Reflective Devices and Associated Equipment; Denial of 
Petition for Rulemaking

AGENCY: National Highway Traffic Safety Administration (NHTSA), DOT.

ACTION: Denial of petition for rulemaking.

-----------------------------------------------------------------------

SUMMARY: This notice denies a petition for rulemaking by Baran Advanced 
Technologies, Ltd., to amend Federal Motor Vehicle Safety Standard No. 
108 to allow the abrupt release of the accelerator pedal to operate 
stop lamps. The reason for the denial is the importance of retaining 
the existing requirement for activating the stop lamps only through 
application of the brake pedal in order to avoid confusion.

FOR FURTHER INFORMATION CONTACT: Patrick Boyd, Office of Vehicle Safety 
Standards, NHTSA (202-366-6346).

SUPPLEMENTARY INFORMATION: Baran Advanced Technologies, Ltd. of Israel 
manufactures a device called ``Red Alert'' which is intended to 
activate the stop lamps during emergency braking before the brake pedal 
is applied, thereby providing an earlier warning for following 
vehicles. The device senses the rate at which the accelerator pedal 
returns to its upper stop after being released. It activates the stop 
lamps for one second if the accelerator pedal reaches its upper stop at 
greater than a certain predetermined rate. Its operation is based on 
the assumption that any rapid release of the accelerator pedal is the 
beginning of an emergency braking maneuver, and thus will be 
immediately followed by application of the brake pedal. Application of 
the brake pedal continues to activate the stop lamp in the usual way. 
If the brake pedal is depressed within one second of the release of the 
accelerator, the brake light will remain on at the end of the initial 
one second period of activation and thus provide a steady signal.
    Standard No. 108 requires that stop lamps ``shall be activated upon 
application of the service brakes.'' That requirement has been 
interpreted to mean that the stop lamps are to be activated only upon 
activation of the service brakes. Some years ago, Baran's competitor in 
the Israeli market, ATAT, sought an interpretation of Standard No. 108 
that would allow the aftermarket installation of its similar Advanced 
Brake Light Device (ABLD) in the U.S. In a letter of January 25, 1990, 
to Larry Snowhite, Esq., the agency concluded that accelerator release 
activation by an aftermarket device such as the ABLD would render the 
stop lamps partially inoperative within the meaning of the prohibition 
of 15 U.S.C. 1397(a)(2)(A). Said NHTSA:

    The heart of our concern is that while the standard requires the 
stop lamp to operate in only one particular circumstance, the ABLD 
causes the stop lamp to operate at an earlier time when the lamp is 
supposed to be unlighted. Further, the ABLD's activation of the stop 
lamp indicates only that the operator has released the accelerator. 
It does not necessarily follow that the brake pedal will be applied. 
Under this fact situation, the stop lamps fulfill a purpose other 
than for which they are installed. This can only create the 
potential for confusion and dilution of the effectiveness of the 
stop signal. For the reasons stated above, we have concluded that 
installation of the ABLD in the aftermarket would render the stop 
lamps partially inoperative.

    Baran's ``Red Alert'' and ATAT's ABLD operate under the same 
fundamental principle of measuring the accelerator pedal return rate to 
anticipate emergency braking. ATAT did not report its threshold rate 
for accelerator pedal release. However, the fact that its false alarm 
rate is similar to that of the Baran system suggests that the 
activation of both systems is based on a similar threshold rate of 
accelerator release. These rates are discussed later in the notice.
    NHTSA has examined Baran's petition in an effort to balance 
possible safety benefits of the device against the possible safety 
disbenefits of signal confusion.
    The first issue is whether the petitioner can demonstrate that its 
device is likely to lessen the number of accidents. Baran estimates 
large reductions in rear-end accidents using figures from a paper by 
Enke ``Possibilities for Improving Safety Within the Driver-Vehicle 
Environment Control Loop'' and from a NHTSA report on Intelligent 
Vehicle Highway System (IVHS) countermeasures to rear end accidents 
(DOT HS 807 995). Enke's paper estimates that the impact speed of 25 
percent of rear end accidents is no more than 10 km/hr (9 ft/sec) and 
that the amount of distance traveled at the assumed initial speed for 
0.25 second equals the distance required to stop from the impact speed. 
Baran claims that ``based on Enke's analysis, providing a driver with 
an additional .25 seconds of warning of an impending stop by the 
leading driver could result in a 25 to 30 percent reduction in all 
rear-end accidents.''
    NHTSA disputes this conclusion. Enke's analysis presumes that the 
following driver is attentive to the very first glimmer of a stop 
signal from the car ahead and that (s)he reacts immediately even though 
a speed differential has not yet occurred. It also presumes that the 
lead driver decelerates so rapidly that the following driver cannot 
``outbrake'' the lead driver and that the initial distance between them 
is less than following driver's reaction time multiplied by the initial 
speed. NHTSA views these presumptions as unrealistic.
    Further, NHTSA's IVHS report characterizes a rear-end crash as 
``largely a dry/straight road phenomenon associated with driver 
inattention.'' It found the lead vehicle stopped (LVS) in 75 percent of 
rear-end crashes, which

    Typically do not involve simply a ``too-slow'' reaction of the 
following driver to a sudden crash threat. In the most common 
scenario, the lead vehicle is stopped for an extended interval 
(i.e., 2-6 seconds) before it is struck by the following vehicle. 
There is adequate time to provide a warning to the following driver 
and for the driver to avoid the crash. Vehicles involved in this 
crash subtype should not be viewed as a locked pair where one 
vehicle is following the other at a specified distance. Instead, the 
following vehicle is closing on a stationary object. The initial gap 
distance between the vehicles is often several hundred feet or more. 
No cases were identified where a lead vehicle decelerated rapidly 
and then was hit by a closely following vehicle immediately after 
coming to a stop.

    It is difficult to see how a 0.25 second advance stop lamp warning 
would be of any use in the 75 percent of rear-end accidents in the LVS 
category when prolonged driver inattention appears to be a near 
universal cause.
    The report characterized the remaining 25 percent of rear-end 
accidents, those in which the lead vehicle was moving (LVM) at impact, 
as follows:

    In contrast, the LVM crash subtype may involve driver reaction 
time following a sudden crash threat as a critical factor. Vehicles 
involved in this circumstance are often ``locked pairs'' with one 
vehicle following the other. However, gaps or following distances 
can range from a few lengths to very substantial distances even in 
this subtype. Not all LVM crashes are precipitated by rapid 
deceleration of the lead vehicle. Many involve slow decelerations 
(e.g., typical slowing before a turn) or simply a speed differential 
between the lead and following vehicles.

    This also indicates that driver inattentiveness is critical in LVM 
crashes. It is hard to accept that a 0.25 second advance in stop lamp 
activation would be of value to an inattentive following driver. NHTSA 
accepts the IVHS report as an accurate reflection of the conditions 
under which rear-end collisions happen, and has concluded that Red 
Alert would not provide an adequate warning to avoid these collisions 
in virtually all the circumstances under which they occur.
    Nevertheless, a manufacturer should not be precluded from offering 
its product, even if safety benefits cannot be demonstrated, unless 
there are potential safety disbenefits created by the product. The 
agency objected to ATAT's ABLD because it was not an unambiguous signal 
of brake application. NHTSA believed that such devices created the 
potential for confusion and dilution of the effectiveness of the stop 
signal. It pointed out that the activation of the stop signal under the 
requested circumstances only signified that the accelerator had been 
released. It did not necessarily follow that the brakes would be 
applied. The brakes were not applied 28 percent of time that the ABLD 
activated in a test report submitted by its proponent, ATAT. Similarly, 
the brakes were not applied 23 percent of the time the ``Red Alert'' 
activated in a test report submitted as part of Baran's petition.
    However, both devices activated the stop lamps far less frequently 
than did the ordinary brake switch activated by depression of the brake 
pedal. During their respective tests, the ABLD activated the stop 
signal about 3 percent as frequently as the ordinary brake switch did. 
``Red Alert'' activated the signal about 1.2 percent as frequently. 
Both tests compared the number of false alarms to the number of 
ordinary stop signal activations of one second or less. False alarms of 
``Red Alert'' were about 2.4 percent as frequent as short brake 
applications, and false alarms of the ATAT device were reported as less 
than 10 percent as frequent as short brake applications. The conditions 
under which the two devices were tested differed greatly. ATAT used a 
special test course, three test vehicles and a limited number of test 
subjects who knew that some undisclosed part of their driving behavior 
was being evaluated. Baran installed ``Red Alert'' on six communal car 
pool vehicles that were operated for a period of months in ordinary 
traffic by various drivers who had no knowledge that their behavior was 
being measured. Given the large variations in activation and false 
alarm rates between test vehicles within each test and the differences 
in test conditions between the ABLD and ``Red Alert'' tests, there is 
no reason to believe that the two devices vary significantly in 
activation rate and false alarm performance.
    The activation rate observed in both tests seems to be too high for 
true emergency braking actions, as neither test documented an actual 
incident of emergency braking. ATAT did not report either observing 
emergency maneuvers or questioning its subjects regarding such 
instances, but it did measure the foot movement times when the 
accelerator control activated the stop lamp. A range between 0.23 and 
0.77 second was observed. When ATAT's subjects were asked to perform 
fast accelerator to brake movements, the 5th and 95th percentile times 
were 0.10 and 0.28 second respectively. It appears that very few of the 
activations in ATAT's test of an accelerator controlled stop lamp were 
rapid enough to indicate an emergency. Baran did not report 
measurements of foot movement times, but it assumed that ``Red Alert'' 
activations were equivalent to instances of emergency braking. ``Red 
Alert'' was characterized as providing an advance warning of 0.35 
second which falls outside of ATAT's rapid foot movement range.
    Both manufacturers argued that the false alarms were insignificant 
because they were few in comparison to the quite large number of brake 
applications of less than one second. Baran also argued that the short 
brake applications themselves had little significance because the 
duration of brake effort would be further reduced by the time consumed 
to build up pressure and to release the brake pedal within one second. 
Baran pointed out that a short brake application may represent only a 
release of cruise control, a state of preparedness or a warning to 
other drivers. While these facts tend to diminish the importance of the 
systematic false alarms, they also demonstrate why following drivers 
are unlikely to act at the instant of the stop lamp illumination in the 
absence of closing speed or some other cue. All the test experience 
reported for accelerator controlled stop lamps involved leading and 
following drivers who were unaware of the use of the device. But this 
device facilitates intentional false alarms as well as systemic 
occurrences. Its widespread use would raise suspicions of intentional 
false alarms on the part of drivers in following vehicles. Intentional 
false alarms can and will occur, sending a misleading signal to the 
driver behind.
    To sum up, the idea that a slightly anticipatory brake lamp will 
prevent large numbers of rear-end collisions is intuitively attractive. 
However, it ignores the fact that lack of driver attention is the root 
cause of rear-end accidents. It is unlikely than an extra quarter 
second of brake light activation would cure or offset the 
inattentiveness of a following driver.
    The agency concludes that the potential safety benefits are 
minimal, but it will consider the results of the Israeli Highway Safety 
Administration's trial of accelerator-controlled stop lamps when they 
are available. While Baran's data indicate that the absolute numbers of 
systematic false alarms may be small on average, a perception by 
following drivers that the brake signal allowed false alarms, whether 
or not intentional, might dilute the currently unambiguous message of 
the stop signal. The agency previously found that ATAT's device would 
render the stop lamp partially inoperative as a result of the same loss 
of unambiguous operation. There are no fundamental differences between 
the systems to cause the agency to change its determination or even to 
allow the agency to distinguish between them in a regulation.
    Accordingly, NHTSA has conducted and concluded a technical review 
of the petition, and has determined that there is not a reasonable 
possibility that the amendment requested would be issued at the 
conclusion of a rulemaking proceeding. Accordingly, the petition by 
Baran Advanced Technologies, Ltd., for rulemaking to amend Standard No. 
108 is denied.

    Authority: 49 U.S.C. 30103; delegation of authority at 49 CFR 
1.50 and 501.8.

    Issued on: July 28, 1994.
Barry Felrice,
Associate Administrator for Rulemaking.
[FR Doc. 94-18802 Filed 8-2-94; 8:45 am]
BILLING CODE 4910-59-P