Immersion and Invariance Control for Reference Trajectory Tracking of Autonomous Vehicles

Gilles Tagne (1) , Reine Talj (1) , Ali Charara (1)

1 Heudiasyc - Heuristique et Diagnostic des Systèmes Complexes [Compiègne]

This paper presents design and validation of a vehicle lateral controller for autonomous trajectory following based on Immersion and Invariance (I&I) principle. The aim is to minimize the lateral displacement of the autonomous vehicle with respect to a given reference trajectory. The control input is the steering angle and the output is the lateral displacement error. A first version of this controller based on I&I principle has been proposed previously by the authors in [1], but it suffers from a lack of robustness in some critical situations, near the nonlinear zones. In this paper, we propose a considerable improvement in the used model and the control synthesis. To validate the control strategy, the closed-loop system simulated on Matlab-Simulink has been compared to the experimental data acquired on the vehicle DYNA of Heudiasyc laboratory, a Peugeot 308, according to several real driving scenarios. The validation shows robustness and good performances of the proposed control approach, and puts in evidence the improvement with regard to the previous I&I control strategy proposed in [1].

Domaines: Automatique

Gilles Tagne, Reine Talj, Ali Charara. Immersion and Invariance Control for Reference Trajectory Tracking of Autonomous Vehicles. IEEE International Conference on Intelligent Transportation Systems (ITSC 2013), Oct 2013, The Hague, Netherlands. pp.2322-2328. ⟨hal-00880307⟩