Human Performance Modeling: Representation of Different Steering Styles

A driver model has been created in order to aid the development of new technologies that have the potential to enhance vehicle handling. This paper describes an investigation into the representation of different steering styles for human performance modeling. Different steering styles result in individual drivers using different steering inputs when negotiating an identical manoeuvre. The work is motivated by the effect of different steering styles on drivers' assessments of vehicles and the consequent possibility of engineering future vehicles to optimise the driver/vehicle combination. To achieve this optimisation, a driver model that is able to digitally represent different steering styles is required. Optimal control theory is used to formulate such a driver model; a cost functional represents the driver's motivation. It is demonstrated that assigning different relative costs to driver effort and vehicle motion results in the driver model using different steering inputs to negotiate a course. Vehicle test results show that different drivers experience different vehicle motion during the same manoeuvre; this is also apparent in results from the driver model.