A driving simulator for road lighting using fixed low mounting height luminaires

摘要

Lighting at grazing incidence is a new concept in so far as this type of road lighting assumes low mounting heights of luminaires installed laterally under the driver's eyes to avoid or limit drastically glare. At the same time, the ratio of spacing to mounting height, even reaching important values compared to conventional lighting solution on poles ,leads to a spacing of 8m to 15m. The consequence is a frequency of "flicker" effect already known in tunnel lighting that is amplified by the speed of vehicles when passing in front of each luminaire.Objective and necessity of a simulator: this kind of wave effect is due to the entrance of light inside the car housing either directly from the luminaire or by reflection of light on the road which is more important under the luminaire. CIE Publication n°88 on tunnel lighting gives some advice about the critical frequencies of the phenomenon to be avoided but this applies to the general case of tunnel where the luminaires are situated on the ceiling or at the top of the "walls". The case of grazing light incidence and the position of the luminaires nearer to the visual axis has not been really studied. At the same time, the checking on site of the possible global stroboscopic effect firstly would suppose quite a long distance of installation (several kilometres) and, secondly, should be based on real driving conditions in which the test can be only subjective for safety reasons. Conscious of the lack of practical knowledge on flicker we decided to create at Thorn in France a simplified driving simulator. Simplified for what concerns the ambient conditions of noises and shaking during the travel but very realistic for what concerns the visual conditions by night, the field of view and the driving task.This paper aims to present the principle of the simulator, the main details and a video recording (3D image synthesis) showing the dynamical effect of the simulated driving task. The latter being the result of a combination of projected images on a large screen placed in front of the windscreen of a real car housing in which the driver is in place. On one side of the car is placed a real luminaire from which are emitted variable luminous intensities directly related to the simulated speed of the vehicle and the dynamical perspective of relative moving luminaires. The drivers uses pedals and the steering wheel to run virtually the car on a one way carriageway comprising two lanes plus a hard shoulder typical of high speed roads. Such a system allows several scenarios of tests using other simulated vehicles moving on the road, fixed (but virtual!) obstacles, slip roads and a service station lit in the vicinity. It is envisaged to record the reaction time of the driver to unexpected events (targets or vehicle to be seen in due time).Obviously these scenarios have been created with the idea to put the observer in driving situation more than visually speaking but also in order to ask him his feelings and the impression he got from the lighting installation for a certain period (circuit of 11km) just after the test. The system is very new so we have only preliminary results of tests from different trials with some observers that will be discussed.