The State of the Art of Driver Model Development 2011-01-0432

The following paper presents an outline of the current state of driver modeling along with the various methods that are employed in their development. In recent years, vehicle manufacturers have implemented various systems that, in some manner, improve the operation of their vehicles. Many of these systems include an electronically controlled device which is capable of making decisions based on the immediate conditions affecting the vehicle. Much of the influence to develop such systems stems from the issue of safety: in emergency situations the control device is capable of making a decision quicker than the driver and thus reduces the potential for some form of collision. Another motivating factor behind these systems is to improve fuel efficiency, specifically in regard to hybrid vehicles where more than one form of propulsion is used and such devices can aid the driver to operate in a more efficient manner. Although these systems have the potential to offer great benefits, in order to be able to implement them, they must function properly such that they provide assistance when necessary while at the same time not being overly intrusive. To achieve this goal, research into driver modeling is gaining more attention in the automotive industry so that driver-in-the-loop (DIL) control systems can be developed. DIL systems work to interpret the driver's intention in order to make a decision whether or not to intervene. The end result of these DIL systems is improved vehicle control without making the driver feel as if their control has been compromised.