Modeling Driver Behavior in a Cognitive Architecture

Salvucci, Dario D.

doi:10.1518/001872006777724417

Cited by 445 publications

(341 citation statements)

References 48 publications

Supporting

Mentioning

334

Contrasting

Unclassified

Order By: Relevance

“…[7] constructed a driver's cognition and action model by using ACT-R, which is a cognitive architecture. [8] presented a model of the driver's cognitive load using resource-allocation theory and a qualitative model.…”

Section: Related Workmentioning

confidence: 99%

Classifying Cognitive Load and Driving Situation with Machine Learning

Yoshida¹,

Ohwada²,

Mizoguchi³

et al. 2014

IJMLC

View full text Add to dashboard Cite

Abstract-This paper classifies a driver's cognitive state in real driving situations to improve the in-vehicle information service that judges a user's cognitive load and driving situation. We measure the driver's eye movement and collect driving sensor data such as braking, acceleration, and steering angles that are used to classify the driver's state. A set of data about the driver's degree of cognitive load, regarded as a training set, is obtained from steering operation and task cognition. Given such information, we use a machine-learning method to classify the driver's cognitive load. We achieved reasonable accuracy in certain driving situations in which the driver moves abnormally for an appropriate service supporting safe driving.Index Terms-Driver's cognitive load, eye movement, machine learning, driving task.

show abstract

Section: Related Workmentioning

confidence: 99%

Classifying Cognitive Load and Driving Situation with Machine Learning

Yoshida¹,

Ohwada²,

Mizoguchi³

et al. 2014

IJMLC

View full text Add to dashboard Cite

show abstract

“…Developed and empirically validated by psychologists, CAs (such as ACT-R) are production-rule based blueprints of the operational structure of cognition [18]. Production rules have been prepared for specific interaction tasks with specific products, especially for use processes that involve typical frequently occurring tasks, in particular driving cars [19], piloting aircraft [20], and interacting with computers [21]. For other 'atypical' tasks related to the use of other products, new production rules need to developed and validated.…”

Section: Related Workmentioning

confidence: 99%

Theoretical underpinning and prototype implementation of scenario bundle-based logical control for simulation of human–artifact interaction

Vegte

Horváth

2012

Computer-Aided Design

View full text Add to dashboard Cite

This article presents a new methodology that enables designers to include in simulations not only the physics aspects of artifact behavior, but also human actions. The motivation for this research came from the fact that none of the conventional approaches to engineering simulations includes manipulative control of products by users as foreseen by designers. By implementing control over physics simulations, changes in parameters can be introduced that alter the course of the simulated process. As a means to do this, we propose to use scenario bundles, with which designers can operationalize their conjectures of how human users interact with products as a series of interconnected simulations. For the imaginary use process described in a scenario bundle, the designer can specify various product designs, user characteristics and environments, which may in each case lead to different concatenations of simulation actions. The proposal facilitates the exploration of possible mismatches and anomalies in use processes. In this article we have described the theoretical fundamentals and the overall concept of the proposed methodology, as well as its realization as a proof-of-concept implementation. This implementation can be used as a tool to specify scenario bundles and to perform controlled simulations of human-product interaction. The use of the tool is demonstrated through a practical example. Although the implementation has proven to be successful in terms of executing scenario bundles, two bottlenecks need further attention: (i) devising stable algorithms for large deformations in physical interaction simulation and (ii) incorporation of already existing algorithms for simulation of low-level human motion control.

show abstract

“…[3], [7] and [2]). These measures can be categorized in three types: However, variability of pilot behavior has to be considered during cockpit design, because the design has to be safe and usable for different types of pilots.…”

Section: Related Workmentioning

confidence: 99%

“…[7] and [3]. Combining mean (as measure of central tendency) and standard deviation (as measure of dispersion) allows to quantify the average case and variability of human and model behavior based on given data samples.…”

Section: Related Workmentioning

confidence: 99%

The Difference Matters: Benchmarking Visual Performance of a Cognitive Pilot Model

Frische

Lüdtke

2011

Human-Computer Interaction – INTERACT 2011

View full text Add to dashboard Cite

Abstract. In this paper we introduce an approach to objectively validate visual performance of a cognitive pilot model using benchmarks of human performance. A study with 16 human airline pilots and two competing models has been conducted in order to validate visual performance of the models applying these benchmarks. The study shows that human performance benchmarks can support analysts with a powerful and easy to use method for validation of human performance models. The benchmark is part of a largerscale method, which will be developed in order to evaluate human factors issues of future HCI-concepts in early stages of system design.

show abstract

Modeling Driver Behavior in a Cognitive Architecture

Abstract: The model can also serve as a core computational engine for practical applications that predict and recognize driver behavior and distraction.

Cited by 445 publications

References 48 publications

Classifying Cognitive Load and Driving Situation with Machine Learning

Classifying Cognitive Load and Driving Situation with Machine Learning

Theoretical underpinning and prototype implementation of scenario bundle-based logical control for simulation of human–artifact interaction

The Difference Matters: Benchmarking Visual Performance of a Cognitive Pilot Model

Contact Info

Product

Resources

About