Skip to Main content Skip to Navigation

Modélisation et contrôle d'un véhicule tout-terrain à deux trains directeurs

Abstract : Interest in autonomous vehicles has increased steadily over recent years. Autonomous driving on rough terrain or with low grip conditions are still technological challenges to be met. The control system must be able to identify the variability of these terrains while following a trajectory. The work described in this thesis focuses on the development of a control strategy to address the problem of autonomous driving with four steering wheels in an off-road environment. In such environments, at high speed, tires are subject to significant slippage, which causes the phenomenon of understeering or oversteering. Thus, the objective is to evolve as quickly as possible in difficult environments while following a reference path whatever the grip conditions. To do so, several points were addressed, such as vehicle and tires modeling, estimation of tire forces and control of four-wheel steering vehicles. The main contributions are: a lateral forces estimation algorithm obtained from two fundamental parameters: the cornering stiffness coefficient and the maximum lateral friction coefficient and a control algorithm allowing to follow a reference trajectory. The control strategy is based on a dynamic model in order to take into account the interactions between the wheel and the ground. Indeed, at high speeds, vehicles are subject to important dynamic phenomena making the kinematic models unrealistic. Kinematic models are only able to represent relatively slow phenomena and therefore are only suitable for low speed vehicles. The developped control strategy will maintain, as far as possible, the tire's behavior in the linear domain. The second steering axle will add an additional degree of freedom allowing to control the vehicle yaw angle with respect to a reference trajectory. This will allow to control the vehicle even in presence of significant slippage and thus increase the vehicle stability.
Document type :
Complete list of metadata
Contributor : ABES STAR :  Contact
Submitted on : Wednesday, December 1, 2021 - 12:56:07 PM
Last modification on : Friday, December 3, 2021 - 3:05:33 AM
Long-term archiving on: : Wednesday, March 2, 2022 - 7:16:22 PM


Version validated by the jury (STAR)


  • HAL Id : tel-03461466, version 1


Laëtitia Li. Modélisation et contrôle d'un véhicule tout-terrain à deux trains directeurs. Robotique [cs.RO]. Université Paris sciences et lettres, 2021. Français. ⟨NNT : 2021UPSLM032⟩. ⟨tel-03461466⟩



Record views


Files downloads