Perception of time to contact in driving simulators

Perception of time to contact in driving simulators

TNO Human Factors Research Institute
Kappé, B.
Korteling, J.E.

1995

When driving a car it is of vital importance that the driver is able to time his actions accurately. Estimating when objects will be reached (the time to contact or TTC) is fundamental to many driving tasks. Visual information is the drivers primary source of TTC information. The present study investigated the way this information should be modeled in the CGI of a driving simulator, in order to get a realistic performance. In the introduction the two major theoretical approaches, i.e., the direct and the inferen-tial method of tion are analysed. In a previous experiment (Schiff & Detwiler, 1979), in which stationary observers estimated the TTC of approaching objects, no effect of the amount of visual information in the scene (image content or complexity) was found on the estimated TTC. The current experiment extends this previous experiment with moving observers estimating the TTC of stationary and moving objects. Results show that TTC estimates in combination with stationary observers are independent of the complexity of the CGI, but that moving observers do estimate TTC more realistic with increasing scene complexity. From these results it can be deducted that, when estimating TTC, observers use mainly the information in optic flow. However, 'static' depth cues can, by inference, reduce a bias in the estimates due to differences in the angular size of the approaching object. The conclusion is that direct and inferential cues are integrated when estimating TTC.

Bij het schatten van de tijd tot contact van een naderend voorwerp blijken proefpersonen vooral gebruik te maken van (globale) optic flow. In de schatting worden "directe" en "indirecte" cues gecombineerd.

Traffic
Vehicle driving
Simulators
Collisions
Motion perception
Optic flow
Time-to-collision

http://resolver.tudelft.nl/uuid:5c56ff72-0d5d-42d5-aaa4-44877c074e67

8114

1995 A053

TNO, Soesterberg Soesterberg

report