WTI

Making Driving Simulators More Useful for Behavioral Research

Primary Investigator Contact Information

External Project Contact(s)

Report(s)

There are no reports associated with this project.

Project Objective

The overall goal of this project is to refine driving simulator research methods so that findings correspond more closely to on road behavior, in order to make driving simulators useful for roadway design.

Project Abstract

Driving simulators have great potential to enhance road safety, because they provide an opportunity to study driver behavior in response to new or revised roadway design features, prior to construction. However, this potential depends on the ability of behavioral data observed in a driving simulator to predict the true behavior that would be committed in the real world. The objective of this research is to fill this important gap by (1) developing transformations of simulator data so that it matches observed on-road behavior, and (2) identifying the level of simulation fidelity needed to address particular design issues. This is a large scale, three phase research program led by the National Advanced Driving Simulator program at the University of Iowa, on behalf of the FHWA Exploratory Advanced Research Program. Phase 1 starts with input from roadway designers and defines simulator characteristics that may be needed to address these issues. Phase 2 collects data to compare driver performance in a range of simulators to on-road performance. Phase 3 develops model-based transformations for relating simulator data to on-road data. WTI is part of the research team, with task responsibilities that include providing simulator resources and developing validation methodology.

Task Descriptions

1. Develop and Perform Scenario Analysis
   1. The team will define driving scenarios that will yield data related to the selected research issues.
2. Collect Simulator Datasets
   1. Using the standard scenario, the team will execute a series of experiments to collect a number of datasets from all the simulators, as well as at least one set of onroad data. All data will be documented in a standard format to facilitate analysis in subsequent tasks.
3. Develop and Perform Model-Based Transformations
   1. Researchers will develop computational driver models that evaluate how differences in experimental test beds (simulator platforms and on-road vehicles) and driving conditions manifest in differences in observed driver behavior. To test the accuracy of the models, data generated from one testbed will be used to predict driver performance in another.
4. Develop and Perform Simulator Analysis
   1. Researchers will use a literature review and surveys to identify simulator characteristics that should be included across all simulator configurations used in the studies that will be conducted during Phase 2.
5. Develop and Apply Simulator-to-Issue Matching
   1. Using the findings from the previous tasks, the team will match the key characteristics of driving simulators (e.g., visual scene size and features, amount of motion) with specific research issues/questions. Researchers will also conduct a preliminary assessment of the degree to which each simulator configuration can provide valid and reliable answers to each research question.
6. Build and Replicate Scenarios
   1. Researchers will replicate the standard scenario on each platform with identical driver performance measures. Technical teams will visit each simulator to conduct an initial assessment to ensure a satisfactory replication of the visual qualities and other elements of the standard scenario.
7. Define Simulator Scenarios and Planned Comparisons
   1. The team will select a standardized driving scenario that can be replicated across all the simulators to be used in the study.
8. Evaluate method and report results
   1. The team will compile and compare the experimental simulator and on-road data and evaluate the success of the overall project.
9. Identify Research and Design Issues
   1. Researchers will solicit input from the FHWA on important roadway design issues, and compile a prioritized list of research issues for consideration.

Milestones, Dates, Schedule

Student Involvement

False

Relationship to Other Research Projects

False

Technology Transfer Activities

True

Transportation Research Board Keywords

Human Factors, driving simulators, driver behavior, roadway design

Partners

Federal Highway Administration (USDOT); National Advanced Driving Simulator, University of Iowa; Battelle Memorial Institute; Entropy Control, Inc.; William H. Levison Associates