Understanding and Modeling the Human Driver

Abstract

This paper examines the role of the human driver as the primary control element within the traditional driver-vehicle system. Lateral and longitudinal control tasks such as path-following, obstacle avoidance, and headway control are examples of steering and braking activities performed by the human driver. Physical limitations as well as various attributes that make the human driver unique and help to characterize human control behavior are described. Example driver models containing such traits and that are commonly used to predict the performance of the combined driver-vehicle system in lateral and longitudinal control tasks are identified.

Comparing and validating models of driver steering behaviour in collision avoidance and vehicle stabilisation

“…These are interpretable, respectively, as (1) steering roughly as deemed necessary given the linear, low-friction vehicle model (K close to 1) to achieve what seems like an unrealistically large safety margin of about 1.5 -3 m, and (2) steering aiming for a small safety margin of about 0 -0.5 m, but applying a steering about three times larger than what the linear vehicle model predicts would be needed for this purpose. A possible interpretation of these fits is that the drivers adapted to the low friction circumstances, responding to vehicle understeering by applying larger steering angles than they would normally [23,49]. Indeed, a separate, cursory analysis of the avoidance steering in the first, unexpected scenario suggests that while steering was predominantly pulse-like already at this point in the experiment (as implied also by Fig.…”

“…However, a recent review, focusing specifically on models that have been applied in simulation of near-collision situations [20], noted two clear limitations in the literature: (a) With just a few exceptions [21][22][23], new models have been proposed without comparing their behaviour to that of existing, alternative models, making it difficult to know which models to prefer for a given application. (b) Validation of model behaviour against human behaviour data from real or reasonably realistic near-crash situations has been virtually non-existent.…”

A multiplicative human steering control model

“…The model has two parameters: ω c which is the combined human-plant open-loop gain, and τ the neuromuscular lag of the humanoperator. The Crossover Model has been widely used as a benchmark to validate other human-control approaches [14], [15]. For the presented application, the data generated from the virtual human is fitted very well by this model (Fig.…”

Drowsiness estimation from identified driver model

“…Because cars are usually equipped with reliable sensors for measuring such data, implementation costs for the proposed method are low. Identification and analysis of a driver model have been studied (Burnham, Seo, & Bekey, 1974;MacAdam, 2003;Plöechl & Edelmann, 2007). Recent research has focused on computer-aided analysis and design methods that use driver models (Boer, Ward, Manser, & Kuge, 2006;Fujinaga, Tokutake, & Miura, 2007;Miura, Tokutake, & Fukui, 2007;Tokutake, Fujinaga, & Miura, 2008;Tokutake, Miura, & Okubo, 2004;Tokutake & Sato, 2005).…”