A multiple-model structure for a driver's internal model of a nonlinear vehicle has been proposed. Multiple models are generated by linearising the nonlinear tyre characteristic at multiple combinations of front and rear slip angle. The multiple-model structure potentially offers a straightforward way to represent a range of driver expertise. The structure is also amenable to the inclusion of a learning function. A model predictive controller compatible with the multiple-model structure has been derived. The controller can be implemented with various assumptions about the driver's ability to predict the future vehicle path. A 'variable-model' prediction might represent the driver who can predict the future operating points of the nonlinear vehicle. A 'fixed-model' prediction might correspond to a less skilled driver, who can only predict on the basis of the present operating point of the vehicle. A 'single-model' prediction might represent a novice driver with no knowledge of the nonlinear vehicle dynamics. Although the application of MPC to nonlinear vehicle steering control is not novel, this proposal to represent degrees of driver expertise and knowledge is believed to be new. Analysis of the controller performance revealed a reduction in the closed-loop controller bandwidth with increasing tyre saturation and fixed controller gains. A parameter study demonstrated that increasing the controller internal model density, increasing the weights on the path error, and increasing the controller knowledge range all improved the path following accuracy of the controller The results presented in the paper suggest that the multiple-model structure and model predictive controller offer significant potential for representing degrees of driver expertise and knowledge. Future work will attempt to fit the model to measured data from novice and expert drivers.
Application of time-variant predictive control to modelling driver steering skill
Anwendung der zeitvarianten Prädiktivregelung zur Modellierung der Fahrerlenkfähigkeit
Vehicle System Dynamics ; 49 , 4-6 ; 527-559
2011
33 Seiten, 26 Bilder, 4 Tabellen, 33 Quellen
Article (Journal)
English
Application of time-variant predictive control to modelling driver steering skill
Online Contents | 2011
|Application of time-variant predictive control to modelling driver steering skill
Taylor & Francis Verlag | 2011
|Application of time-variant predictive control to modelling driver steering skill
Automotive engineering | 2011
|Switching-Based Stochastic Model Predictive Control Approach for Modeling Driver Steering Skill
Online Contents | 2015
|