CORM: Constrained Optimal Reconfiguration Matrix for Safe On-Ramp Cooperative Merging of Automated Vehicles

Lyes Saidi (1) , Lounis Adouane (1, 2) , Reine Talj (1)

1 Heudiasyc - Heuristique et Diagnostic des Systèmes Complexes [Compiègne]
2 IP - Institut Pascal

On-ramp cooperative merging maneuver is currently one of the most challenging tasks for cooperative and automated vehicles (CAVs). Safety concern is the key aspect to capture the performance of the merging strategy. In this paper, it is proposed to adapt the dynamic inter-target distance matrix originally developed in [13] in order to explicitly integrate the environmental constraints (e.g., road borders). The main objectives of the proposed Constrained Optimal Reconfiguration Matrix (CORM) are to ensure safe and reliable navigation of the CAVs in formation and guarantee a smooth merging maneuver, while taking into account the on-road environment constraints. An analytical model of the formation composed by the CAVs is presented, in addition to a formal demonstration of the respect of the in-between distances using the constrained optimal reconfiguration matrix. Simulations in different scenarios are performed to evaluate the safety and the efficiency of the proposed approach.

Domaines: Automatique / Robotique

Lyes Saidi, Lounis Adouane, Reine Talj. CORM: Constrained Optimal Reconfiguration Matrix for Safe On-Ramp Cooperative Merging of Automated Vehicles. 25th IEEE International Conference on Intelligent Transportation Systems (ITSC 2022), Oct 2022, Macau, China. pp.2783-2790, ⟨10.1109/ITSC55140.2022.9921870⟩. ⟨hal-03872810⟩