Human Factors Human Factors Research Issues for Research Issues for Cooperative Cooperative Intersection Intersection Collision Avoidance Collision Avoidance Systems Systems (CICAS) (CICAS) Vicki Neale, Ph.D. Vicki Neale, Ph.D. Director, Center for Vehicle- Director, Center for Vehicle- Infrastructure Safety Infrastructure Safety Virginia Tech Transportation Institute Virginia Tech Transportation Institute
19
Embed
Human Factors Research Issues for Cooperative Intersection Collision Avoidance Systems (CICAS) Vicki Neale, Ph.D. Director, Center for Vehicle-Infrastructure.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Human Factors Human Factors Research Issues for Research Issues for
Human Factors Research Human Factors Research IssuesIssues
Each warning system has human factors Each warning system has human factors research needs in common, although the method research needs in common, although the method to address the research need and final answer to address the research need and final answer may be different for each.may be different for each. Development of an Algorithm (Warning timing)Development of an Algorithm (Warning timing)
Depends on warning type and driver characteristicsDepends on warning type and driver characteristics Driver alertness (Willful violation vs. Unintentional due to Driver alertness (Willful violation vs. Unintentional due to
distraction)distraction) AgeAge
For example, with Gap acceptance issues, if design for older For example, with Gap acceptance issues, if design for older drivers, younger drivers may consider warning “too early”drivers, younger drivers may consider warning “too early”
Determining the driver interface (in-vehicle or Determining the driver interface (in-vehicle or infrastructure)infrastructure)
Determining acceptance of alert rates, including Determining acceptance of alert rates, including nuisance alarm ratesnuisance alarm rates
Determining scenarios for which warning is Determining scenarios for which warning is appropriateappropriate
Determining overall system reliabilityDetermining overall system reliability
Example: Addressing HF Example: Addressing HF Issues in CICAS-V Issues in CICAS-V
TASK 3.1:100 CAR DATA
MINING
TASK 3.3:SMART ROAD
STUDIES
PREVIOUS RESEARCH:
OEMS
ICAVIDS
OTHER
Timing Corrections
Timing Intervals
PLANNING AND
COORDINATION TASK
TASK 13:FOT
EXPERIMENTAL DESIGN
TASK 10:FOT
PROTOTYPE BUILD
TASK 11:OBJECTIVE
TESTING
TASK 3.4:PILOT FOT
TASK 3.2: NATURALISTIC
DATA COLLECTION
Data to Support USDOT / Volpe
Crash Benefits Estimation
CICAS-V Subtask 3.1CICAS-V Subtask 3.1 Sample Research Question:Sample Research Question: What are the circumstances under which a What are the circumstances under which a
driver violates?driver violates? Willful vs. unintentional violatorWillful vs. unintentional violator
Intersection control type (signal, stop sign) Intersection control type (signal, stop sign) Time after red phase onset (for signalized intersection Time after red phase onset (for signalized intersection
only)only) Visual checking behavior (e.g., looking both ways at Visual checking behavior (e.g., looking both ways at
intersection) intersection) Driver ageDriver age Driver genderDriver gender Driver aggressiveness (e.g., car following behavior)Driver aggressiveness (e.g., car following behavior) Following driver presence/headwayFollowing driver presence/headway Intersection approach speedIntersection approach speed Posted speed limitPosted speed limit Traffic densityTraffic density Time of dayTime of day Weather/Visibility/Road conditionsWeather/Visibility/Road conditions
To Address the Research To Address the Research QsQs
Mining the 100-Car Database to Mining the 100-Car Database to understand what drivers are doing understand what drivers are doing when they commit violations and when they commit violations and near-violations.near-violations.
CICAS-V Subtask 3.2CICAS-V Subtask 3.2
Research QuestionsResearch Questions How many false alarms and misses How many false alarms and misses
result from any particular algorithm?result from any particular algorithm? How often do vehicles violate any How often do vehicles violate any
particular signalized or stop controlled particular signalized or stop controlled intersection?intersection?
Use naturalistic driving approaches Use naturalistic driving approaches to intersections to refine the alert to intersections to refine the alert timing approach and determine rates timing approach and determine rates of violations and near-violationsof violations and near-violations
To Address the Research QsTo Address the Research Qs
Predicted Metric Example
Compliance Violator
Compliance Correct Rejection False Alarm Observed
Violator Miss Hit
CICAS-V Subtask 3.3CICAS-V Subtask 3.3
Determine the Driver Vehicle Determine the Driver Vehicle InterfaceInterface Is the warning meaningful?Is the warning meaningful? Does it illicit the appropriate response Does it illicit the appropriate response
in a timely manner?in a timely manner? DVI Issues DVI Issues
Driver perception of scenarioDriver perception of scenario Driver reaction timeDriver reaction time Final algorithmFinal algorithm
To Address the Research To Address the Research QsQs
Conducting a series of controlled Conducting a series of controlled Smart Road tests to evaluate drivers’ Smart Road tests to evaluate drivers’ response to DVIs in a “surprise” response to DVIs in a “surprise” trial.trial.
Open State(alert issued; light turns from amber to red)
Closed State for 2 Seconds
(light turns amber)
Occlusion technique used to simulate driver distraction and place precise
experimental control over when forward scene and traffic signal phase
can be viewed
VTTI Smart Road intersection
CICAS-V Subtask 3.4CICAS-V Subtask 3.4
How will the final CICAS-V system How will the final CICAS-V system function in the real-world with naïve function in the real-world with naïve drivers?drivers?
What needs to be changed?What needs to be changed? Are we FOT ready?Are we FOT ready?
To Address the Research To Address the Research QsQs
Conduct a pilot FOT with naïve Conduct a pilot FOT with naïve drivers using the final FOT CICAS-V drivers using the final FOT CICAS-V system design.system design.
This final HF task is the culmination This final HF task is the culmination of the coordination of all of the HF of the coordination of all of the HF and non-HF tasks.and non-HF tasks.
Challenges/Next StepsChallenges/Next Steps
Conduct FOT to investigate potential Conduct FOT to investigate potential safety benefits and customer acceptance safety benefits and customer acceptance associated with the systemassociated with the system
Determine how to integrate the CICAS Determine how to integrate the CICAS warnings with each other and with other warnings with each other and with other in-vehicle warningsin-vehicle warnings Understand how integrated systems perform Understand how integrated systems perform
in the real world in the real world Maintaining the necessary coordination Maintaining the necessary coordination