2026-09-18T19:32:25Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/345802022-01-13T07:54:11Zcom_1721.1_7582com_1721.1_7581col_1721.1_131023
Charles M. Oman.
Aboukhalil, Anton
Massachusetts Institute of Technology. Dept. of Aeronautics and Astronautics.
Massachusetts Institute of Technology. Department of Aeronautics and Astronautics
2006-11-07T13:07:51Z
2006-11-07T13:07:51Z
2006
2006
http://hdl.handle.net/1721.1/34580
71225425
Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 2006.
Includes bibliographical references (p. 145-148).
Falling asleep while operating a vehicle leads to serious accidents and loss of lives. The challenge of detecting drowsiness nonintrusively stems from the absence of a single marker, and the existence of diverse signs and symptoms that collectively but not uniquely characterize it. Current alerters for locomotive cabs are inadequate partly because they a) often monitor one modality b) fail to consider the inherent characteristics of the locomotive operator tasks and physical environment, and c) are typically developed without quantitative techniques to assess performance and optimize components as part of an overall system, rather than at the individual level. Based on an estimation theory framework, a new systems approach is here proposed to design locomotive cab alerting technologies. The main idea is to combine information from an infrared eyelid monitor and a generic Train Sentry class activity monitor, to isolate the common drowsiness component and obtain an improved estimate of the operator's state. A study first quantified the important physical aspects of the locomotive cab and engineer behavior pertinent to the performance of image-based eye closure monitors.
(cont.) A bench test study evaluated Attention Technology's current infrared eye closure monitor prototype front-end image analysis, Model DD-850, to verify whether its performance was a good match to the locomotive physical environment and engineer behavioral characteristics. Data from these studies were used to develop a simulation software tool in MATLAB/SIMULINK. The goal was to assess the proposed tandem detector solution and to support rational design, development and optimization of future locomotive alerting systems. A signal detection theory (SDT) approach was employed. However, the detectors were nonlinear, had multiple alerting levels and displayed non-Gaussian noise characteristics. Therefore, Monte Carlo methods were used to compute their SDT parameters on both a standalone and tandem basis. Investigation through simulation showed that adopting an architecture using tandem detectors and an "AND" logic based arbiter reduces the false alarm rate by an order of magnitude and improves the total time to alert, at the expense of only a few percent in missed alarm probability. Detection performance may be further enhanced using a speed dependent arbiter with "AND" logic above a speed threshold and "OR" logic below it.
(cont.) In the simulation, the speed threshold was found to be 25 mph. This system-level study provides enough ground to build a prototype system and test the proposed solution in a simulator. Supported by FRA through Department of Transportation Volpe Research Center Contract DTRS57-04-Q-80164 PR 79-3354 and by a Postgraduate Scholarship from the Natural Sciences and Engineering Research Council of Canada.
by Anton Aboukhalil.
S.M.
148 p.
8248380 bytes
8256477 bytes
application/pdf
application/pdf
eng
Massachusetts Institute of Technology
CDROM contains files in MATLAB v.7.0.4 and SIMULINK 6.2.1.
M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission.
http://dspace.mit.edu/handle/1721.1/7582
Aeronautics and Astronautics.
Systems approach to the design of locomotive fatigue management technologies
Thesis
application/pdf
false
Publication
unknown
unknown
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="766196bd-5c08-47c7-830c-fc28ca3678fe">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Language>eng</Language>
<Title>Systems approach to the design of locomotive fatigue management technologies</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2006</PublicationDate>
<Authors>
<Author>
<DisplayName>Aboukhalil, Anton</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<DisplayName>Massachusetts Institute of Technology</DisplayName>
<OrgUnit />
</Publisher>
</Publishers>
<License>http://dspace.mit.edu/handle/1721.1/7582</License>
<Keyword>Aeronautics and Astronautics.</Keyword>
<Abstract>Falling asleep while operating a vehicle leads to serious accidents and loss of lives. The challenge of detecting drowsiness nonintrusively stems from the absence of a single marker, and the existence of diverse signs and symptoms that collectively but not uniquely characterize it. Current alerters for locomotive cabs are inadequate partly because they a) often monitor one modality b) fail to consider the inherent characteristics of the locomotive operator tasks and physical environment, and c) are typically developed without quantitative techniques to assess performance and optimize components as part of an overall system, rather than at the individual level. Based on an estimation theory framework, a new systems approach is here proposed to design locomotive cab alerting technologies. The main idea is to combine information from an infrared eyelid monitor and a generic Train Sentry class activity monitor, to isolate the common drowsiness component and obtain an improved estimate of the operator's state. A study first quantified the important physical aspects of the locomotive cab and engineer behavior pertinent to the performance of image-based eye closure monitors.</Abstract>
<Abstract>(cont.) A bench test study evaluated Attention Technology's current infrared eye closure monitor prototype front-end image analysis, Model DD-850, to verify whether its performance was a good match to the locomotive physical environment and engineer behavioral characteristics. Data from these studies were used to develop a simulation software tool in MATLAB/SIMULINK. The goal was to assess the proposed tandem detector solution and to support rational design, development and optimization of future locomotive alerting systems. A signal detection theory (SDT) approach was employed. However, the detectors were nonlinear, had multiple alerting levels and displayed non-Gaussian noise characteristics. Therefore, Monte Carlo methods were used to compute their SDT parameters on both a standalone and tandem basis. Investigation through simulation showed that adopting an architecture using tandem detectors and an "AND" logic based arbiter reduces the false alarm rate by an order of magnitude and improves the total time to alert, at the expense of only a few percent in missed alarm probability. Detection performance may be further enhanced using a speed dependent arbiter with "AND" logic above a speed threshold and "OR" logic below it.</Abstract>
<Abstract>(cont.) In the simulation, the speed threshold was found to be 25 mph. This system-level study provides enough ground to build a prototype system and test the proposed solution in a simulator. Supported by FRA through Department of Transportation Volpe Research Center Contract DTRS57-04-Q-80164 PR 79-3354 and by a Postgraduate Scholarship from the Natural Sciences and Engineering Research Council of Canada.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>