Driver Adaptation Research Articles

Modeling driver steering behavior in restricted-preview boundary-avoidance tasks

In the design of human-like steering support systems, driver models are essential for matching the supporting automation's behavior to that of the human driver. However, current driver models are very limited in capturing the driver's adaptation to key task variables such as road width and visibility (i.e., ‘preview’ of the road ahead). This paper uses a recently proposed, novel control-theoretical model for centerline tracking to investigate driver steering in lane-keeping tasks with restricted and unrestricted preview, in an attempt to substantially extend this model's validity. Using data from a tailored driving simulator experiment, three driver control loops (feedforward, heading and position feedback) are separately quantified using system identification techniques. The results show that when preview is restricted, drivers use all of the remaining preview to anticipate the curves of the road ahead, and are no longer able to ‘smooth’ tight curves in the road trajectory (i.e., corner cutting). When sufficient preview and lane width are available, the time to line crossing increases, and steering behavior is less aggressive and more intermittent, or more ‘satisficing’. The novel driver steering model captures these adaptations very well (over 95% of the steering actions) and can thereby be instrumental in realizing human-like steering automation and support systems.

Investigations on the required luminance in tunnel entrances

The method described to derive the luminance in the entrance zone of tunnels has been based on physiological considerations of the adaptational processes taking place when approaching a tunnel. The driver's adaptation can be sufficiently approximated by the luminance prevailing in the entrance he is fixating and the superimposed equivalent veiling luminance Lseq caused by the surrounding field. The required luminance level in the entrance zone can be obtained by postulating that a 10' object of a defined contrast should be visible as the straylight reduces the contrast in the eye of the approaching driver. Additional investigations were carried out in which the observer had to evaluate subjectively the entrance luminance in simulated tunnel scenes because it was uncertain what contrast of the object could serve to sufficiently describe the visual task. It was shown that levels rated to provide between fair and good safety conditions almost coincided with the visibility of 10' target with C = 0.25. With this criterion the required entrance luminance L2 results as a rule of thumb to be twice the measured or calculated veiling luminance Lseq. A decrease of safety demands to "fair" nearly corresponding to C = 0.3 would allow L2 to be reduced to two-thirds of that value.

A Driver Glare-Discomfort Model to Evaluate Automotive Stop Lamp Brightness

Laboratory and field studies were conducted to explore the applicability of previously reported laboratory-based discomfort glare threshold data to the prediction of the discomfort level drivers may experience due to glare from the stop lamps of a lead vehicle. The results show that the BCD (Borderline between Comfort and Discomfort) threshold brightness data developed by Putnam and Faucett (1951)* can be used to predict various discomfort levels experienced by the driver under field conditions. A prototype glare-discomfort model was developed which predicts driver discomfort as a function of distance between the two vehicles, lamp intensity, size, and the driver's adaptation to ambient luminance levels. The model incorporates a threshold multiplier to account for differences between the BCD thresholds and field-observed glare-discomfort ratings due to glare-exposure duration and multiple-glare sources.

Identification of Steady Burning Red Lamps as Tail or Stop Signal in Night Driving

A three phase investigation was conducted to assess the probabilities of identifying steady burning red signals as tail or stop lamps as a function of their intensity and size and the driver's adaptation brightness.

Experiment on the Visibility of the Tunnel Entrance Lighting

One of the most serious problems connected with tunnel lighting is the so-called “black hole effect” for long tunnel and the “black frame effect” for short tunnel. To eliminate these “black hole and frame effects”, a sufficient luminance has to be given to the road surface and walls in the tunnel.Previous investigators, such as Waldram5) , Brass6) , de Boer9) and Schreuder2) revealed that the necessary values of luminance are at least 10 to 12.5 % for driver's adaptation luminance. Some authors of previous studies on the tunnel entrance lighting stated that the outdoor luminance near tunnel can be regarded as the adaptation luminance of the driver.However, the adaptation luminance of human eye is not determined by the average luminance of the visual field, but strongly influenced by the luminace projected upon the fovea of retina, so the adaptation luminance of the driver approaching the tunnel mouth will be gradually reduced.It is also clear from this consideration that high luminance of tunnel exit projected upon the driver's central vision will give a contrary effect on the adaptation. Therefore, the adaptation of the driver approaching the tunnel will differ for different distances from the tunnel and apparent lengths of the tunnel for the driver.The visual condition of the driver on the motor vehicle is carefully analyzed. Previous investigators stated10) 16) that the angular velocity of the object under 5 degrees per second gives a negligible influence on the ability of the eye. The angular velocity of objects, which is important for the driver, on the road some distance ahead and observed from a high speed vehicle is only several tens minutes per second, even the speed of the motor vehicle being 80 km/h or more. Thus, it is found that the influence of the angular velocity of the object is negligible, but the object must be perceived within a very short time, because the driver has to gather a lot of information at each instance.With this born in mind, a simulator has been constructed which reproduces the visual condition of the driver approaching the tunnel as faithfully as possible (Fig. 4).The observer looked at a simulated tunnel entrance which was extended perspectively with a synchronous motor driven on inclined rails (Fig. 5) reproducing the change of the apparent size of the tunnel mouth as looked from the driver who was driving a motor vehicle on the road leading to the tunnel. The observer pre-adapted a bright, uniform visual field having a luminance of 4000 cd/m2 for 15 seconds. In this condition “the outside of the tunnel” and “the inside of the tunnel” of the simulator were-brightened. After 15 sec. of pre-adaptation, the simulated tunnel entrance began to be extended, and the lighting of only “the inside of the tunnel” was suddenly switched off, although luminance to be tested was kept in the tunnel. Eight differnt signals were sent to the observer, they were formed by different combinations of exposure position of the three rectangular objects with an apparent size of 15×15 cm2 as viewed at a distance of 100 m and luminance contrast of 25 % with regard to the bac-kground, the objects being barker than the background. The observer perceives the “position” of the objects exposed, and operated push button switches corresponding to the positions of the object perceived.The time of the object exposed was 0.5 sec. in one observation. All the combinations of the positions of the object exposed and the luminance in the tunnel tested were presented to the observer in a random order.Two different observer groups took part in the experiments, one for long tunnels and the other for various lengths of short tunnels.