A three level trajectory planning for autonomous transportation systems

Duc Thang Nguyen (1) , Hadi Zaatiti (1) , Mohamed Tlig (1) , Jean van Frank (1)

1 IRT SystemX

Autonomous vehicles are considered as means of transport for the future. However, the design and the development of autonomous systems such as public transport, robot taxi and good delivery are challenging. On one hand, it involves the maturity of autonomous vehicles. On the other hand, it relates to the management of these new services. One of the major considerations for these systems is how to safely navigate autonomous vehicles to fulfil their missions. In this paper, we propose a method for navigating autonomous vehicles through a three level trajectory planning. This method includes: 1) the computation of a global trajectory for each vehicle, 2) a guidance for hundreds of meters in front of the vehicle and 3) a real time computation for an immediate trajectory based on the current situation. This method provides the guidance which is easy to interpret by the autonomous vehicles for their decision making so that the safe navigation is insured. Additionally, the approach reveals to be flexible for incorporating trajectory adjustments.

Mots clés: Autonomous vehicle; navigation; trajectory planning; public transportation

Domaines: Sciences de l'ingénieur [physics]; Informatique [cs]

Duc Thang Nguyen, Hadi Zaatiti, Mohamed Tlig, Jean van Frank. A three level trajectory planning for autonomous transportation systems. CONGRÈS : SIA Simulation Numérique digital, Apr 2021, En ligne, France. ⟨hal-03202335⟩