Skip to Main content Skip to Navigation

Réduction du mal du simulateur lors de la simulation longue durée pour la validation de véhicules autonomes

Abstract : Autonomous vehicles are considered to be the major innovation vehicle of the comingdecades. It is therefore crucial to be able to assess and validate upstream the complexintelligent systems involved in these future vehicles as well as their interactions with thedriver. This is why simulation tools are favored in order to be able to carry out thebillions of kilometers of driving required, as well as critical cases, without endangeringusers. The driving simulation driver-in-the-loop will be particularly useful in order tostudy and validate the phases of hybrid driving, such as the delegation or resumption ofdriving by the driver. However, the use of immersive driving simulators is particularlylimited by the simulator sickness. Indeed, this phenomenon, similar to motion sickness,can affect the driver’s performance and task of the driver, as well as limit the simulationtime.This thesis « Reducing simulator sickness during long-term simulation for autonomousvehicle validation » was carried out within the framework of a CIFRE contract betweenArts et Métiers on the one hand and the Renault group on the other hand. First, we wereinterested in understanding this phenomenon through the literature.We were particularlyinterested in its origins as well as the various factors that can influence it. These analyzesallowed us to identify different levers that could allow us to reduce the simulator sickness.We then carry out a series of experiments in order to test and implement our chosensolutions. The first experiment focuses on the effect of adding vibratory modalities of thecabin. The second one applies to correct a latency gap between the dynamics and thevisual. Finally, the last study is interested in the possible triggering of unexpected ocularreflexes by motion control algorithms.The thesis will present the experimental results of the diminution of symptoms ofsimulator sickness through the integration of vibrations or compensation of latencies. Itwill also expose the unexpected triggering of oculomotor mechanisms that can be a sourceof oscillopsias, or even symptoms of motion sickness. Thus allowing avenues for significantimprovements of simulator sickness, in order to conduct autonomous vehicle analysis andvalidation campaigns.
Document type :
Complete list of metadata
Contributor : ABES STAR :  Contact
Submitted on : Thursday, October 28, 2021 - 5:44:57 PM
Last modification on : Friday, August 5, 2022 - 2:54:01 PM
Long-term archiving on: : Saturday, January 29, 2022 - 8:01:47 PM


Version validated by the jury (STAR)


  • HAL Id : tel-03407970, version 1


Guillaume Lucas. Réduction du mal du simulateur lors de la simulation longue durée pour la validation de véhicules autonomes. Traitement du signal et de l'image [eess.SP]. HESAM Université, 2021. Français. ⟨NNT : 2021HESAE006⟩. ⟨tel-03407970⟩



Record views


Files downloads