99-2020. Ocular-based Measures of Driver Alertness; Notice of Conference and Request for Submissions

[Federal Register Volume 64, Number 18 (Thursday, January 28, 1999)]
[Notices]
[Pages 4488-4490]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 99-2020]


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

DEPARTMENT OF TRANSPORTATION

Federal Highway Administration
National Highway Traffic Safety Administration


Ocular-based Measures of Driver Alertness; Notice of Conference 
and Request for Submissions

AGENCY: Federal Highway Administration (FHWA), DOT.

ACTION: Notice of conference and request for submissions.

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

SUMMARY: This notice is both an invitation to participate in a 
conference addressing ocular-based measures of driver alertness and a 
request for submissions to be presented/demonstrated at the conference. 
The conference is being sponsored by the FHWA's Office of Motor 
Carriers and the National Highway Traffic Safety Administration's 
(NHTSA) Office of Vehicle Safety Research (formerly, the Office of 
Crash Avoidance). The purpose of the conference is to (1) share recent 
FHWA/NHTSA findings regarding the validity of eye-based measures of 
driver alertness, (2) share recent FHWA and NHTSA technology 
developments in this area, (3) identify and provide information about 
other Research and Technology (R&T) studies relevant to in-vehicle 
alertness monitoring, (4) review the overall state-of-the-art of in-
vehicle alertness monitoring, (5) review concepts for feedback of 
alertness information to drivers and other proposed features of the 
driver-vehicle interface, and (6) review concepts for the successful 
and user-acceptable introduction of in-vehicle alertness monitoring 
systems to commercial motor carrier fleets.

DATES: The conference will be held on April 26-27, 1999. Each day's 
session will begin at 9 a.m. and end at 5 p.m. Papers and technology 
demonstration proposals/abstracts must be received on or before March 
1, 1999.

ADDRESSES: The conference will be held at the Hyatt-Dulles Hotel, 2300 
Dulles Corner Boulevard, Herndon, Virginia.

FOR FURTHER INFORMATION CONTACT: For conference information and to 
obtain

[[Page 4489]]

registration materials, contact Ms. Annette Smith, Portfolio Management 
Group, Ltd., 8513 Ashwood Drive, Capitol Heights, MD 20743; Telephone: 
[phone]; FAX: [phone]; E-mail: [e-mail]. Paper 
and technology demonstration proposals should be submitted to Robert J. 
Carroll, Office of Motor Carrier Research and Standards (HCS-30), 
Federal Highway Administration, 400 Seventh Street, SW., Washington, DC 
20590; Telephone: [phone]; FAX: [phone]; E-mail: 
[e-mail].

SUPPLEMENTARY INFORMATION: From 1997-1998, the Intelligent 
Transportation Systems/Commercial Vehicle Operations Division of the 
FHWA's Office of Motor Carriers funded a program, which was managed by 
the NHTSA's Office of Crash Avoidance Research, to study competing 
fatigue detection technologies. Under the program, the University of 
Pennsylvania (UPENN) conducted laboratory experiments to evaluate the 
validity, sensitivity, and reliability of selected personal 
(psychophysiological) fatigue-detection devices and measures. The study 
evaluated eye closure measures such as PERCLOS, a measure of eyelid 
droop identified in earlier NHTSA research as being a promising index 
of fatigue. PERCLOS is defined as the percent of time eyelids are 
closed 80% or more--a measure found to be significantly correlated with 
driver fatigue. Other psychophysiological measures assessed included 
two eye-blink measures, two electroencephalograph (EEG) measures, and a 
head movement detector. All measures had some validity, but the results 
corroborated most strongly the validity of PERCLOS. The final report, 
``Evaluation of Techniques for Ocular Measurement,'' DOT-HS-808 762, is 
available through the National Technical Information Service, telephone 
[phone] or [phone]. The FHWA and the NHTSA believe that 
the PERCLOS measure is one of the most promising known real-time 
indicators of driver alertness for in-vehicle systems. Laboratory-based 
studies of the driver-vehicle interface for a PERCLOS-based in-vehicle 
alertness monitoring system are currently underway. This Intelligent 
Vehicle Initiative (IVI)-funded project is being managed by the NHTSA's 
Office of Crash Avoidance Research and is a follow-up to the PERCLOS 
validation study. Under the program, UPENN is conducting laboratory 
experiments to evaluate the effectiveness of various potential elements 
of the driver-vehicle interface (DVI) of in-vehicle CMV driver 
alertness monitoring devices. Also, under this program, Carnegie Mellon 
Research Institute has developed, and is testing, a new camera and 
related software that can monitor and analyze a driver's PERCLOS in 
real-time. Other DVI components to be assessed include real-time 
gauges, informational alarms/warnings, and alerting stimuli. The study 
will make recommendations regarding optimal DVI design elements for CMV 
driver alertness monitors. This new study will be completed in the 
Spring of 1999 and reported at the conference. This technical 
conference/workshop is planned to discuss recent scientific validation 
findings regarding PERCLOS and other eye activity measures as metrics 
of alertness, and the status of efforts to develop in-vehicle sensors 
to continuously measure PERCLOS as an indicator of driver alertness 
(i.e., develop an ``alertometer''). The conference will also address 
the potential and appropriate uses of ``alertometer'' data and ways to 
ensure the active participation and acceptance of drivers and 
management in the use of such technologies. Since the PERCLOS measure 
will likely be a key metric employed in any operational test of 
alertness monitoring technology, the workshop will also provide an 
update of the FHWA's plans for implementing the IVI operational tests, 
as described above.
    Due to limited seating, early registration is encouraged. The 
registration fee is $150. The registration fee for full-time students 
is $100. Those registering before March 1, 1999 may pay an early 
registration fee of $100 ($75 for full-time students). Full refund of 
registration fees will be made for cancellations received by April 15, 
1999. Refunds of 75 percent of the registration fee will be made for 
cancellation notices received after April 15, 1999. Persons interested 
in attending, presenting papers, and/or demonstrating relevant 
technologies are invited to obtain registration materials and submit 
papers or technology demonstration proposals, in accordance with the 
criteria set forth below.
    To be accepted for presentation, papers proposing methods of 
alertness measurement should describe how the measure/method meets one 
of the following three criteria for continuous, in-vehicle driver 
alertness monitoring, regardless of the technology used:
    (1) Provide continuous, in-vehicle driver-alertness monitoring, 
using the PERCLOS measure.
    (2) Be highly correlated with the PERCLOS.
    (3) Have gone through a validation process similar to the PERCLOS 
validation.
    Technologies proposed for demonstration at the conference will be 
evaluated with respect to their validity in measuring driver alertness, 
their reliability/durability for in-vehicle operation, and their 
current or potential cost of production and integration in vehicle 
operations. Interested parties are instructed to request a registration 
packet and submit a 300-500 word abstract, describing the paper or 
technology demonstration proposed for presentation at the conference. 
The abstracts related to technology demonstrations should include a 
description of the technology, including its major components, 
functional basis (i.e., what it measures), how it is installed in a 
vehicle (and/or worn by operators), what kind of feedback it provides 
to drivers, evidence of validity (e.g., correlation with performance on 
vigilance or alertness-related tasks), operational reliability/
durability, unobtrusiveness/acceptability, and affordability (or 
potential affordability following further development). Empirical data 
on device validity (i.e., evidence that it is accurately measuring 
alertness as measured by some independent criterion which is a known 
valid measure of alertness, such as the psychological vigilance test) 
is particularly important. The FHWA and NHTSA will select or invite 
presentations and demonstrations for the conference, based upon these 
criteria. Submitters are instructed not to submit any confidential or 
proprietary data on device design or performance.
    The outcomes of this conference are expected to be (1) greater 
public awareness of recent FHWA/NHTSA findings with respect to the 
validity of eye-based measures of driver alertness and related 
technology developments in this area, (2) the identification of other 
research and technology studies relevant to in-vehicle alertness 
monitoring, (3) a review of the overall state-of-the-art of in-vehicle 
alertness monitoring, (4) a review of concepts for providing feedback 
of alertness information to drivers, as well as other proposed features 
of the driver-vehicle interface, and (5) a review of concepts for 
successful and user-acceptable introduction of in-vehicle alertness 
monitoring to commercial motor carrier fleets. The FHWA does not 
believe the outcomes of this conference will impact the agency's on-
going rulemaking addressing its prescriptive hours-of-service rules.


[[Page 4490]]


    Authority: 23 U.S.C. 315; 49 U.S.C. 31136; 49 U.S.C. 31502; 49 
CFR 1.48 and 1.50.

    Issued on: January 22, 1999.
George L. Reagle,
Associate Administrator for Motor Carriers Federal Highway 
Administration.
Raymond P. Owings,
Associate Administrator for Research and Development, National Highway 
Traffic Safety Administration.
[FR Doc. 99-2020 Filed 1-27-99; 8:45 am]
BILLING CODE 4910-22-P