Name: PEDRO HENRIQUE VIEIRA DE OLIVEIRA AZEVEDO
Publication date: 04/09/2023
Examining board:
Name |
Role |
|---|---|
| ALBERTO FERREIRA DE SOUZA | Coorientador |
| CLAUDINE SANTOS BADUE | Presidente |
| DENIS FERNANDO WOLF | Examinador Externo |
| KARIN SATIE KOMATI | Examinador Externo |
| MARIA CLAUDIA SILVA BOERES | Examinador Interno |
Pages
Summary: The use of autonomous vehicles on public roads and in industrial environments has been growing in recent
years and in the core of autonomous vehicles operation, route planning plays a fundamental role in computing a route
through a road network from the autonomous vehicle’s current pose to a desired final goal. Route planning is a crucial
component of autonomous vehicle operation, and it requires a dynamic representation of the road context. However,
the representations provided by open-source solutions do not match the quality of those provided by private
companies. In this work, we propose a planning pipeline for autonomous vehicle software systems. The pipeline is
composed of an offline process and an online process. The offline process constructs a road network and consists of
two modules: the Waypoints Editor and the Multi-Level Road Network Generator. The online process consists of three
modules: the Route Planner module to compute routes, the Off-Road Planner module to paths in order to bring the
car to the start of the route or to take the car from the end of the route to the final goal, and the Frenét Frames Path
Planner to generate alternative paths to the right and left of the route and, in the presence of static obstacles, overtake
them. We evaluated the performance of the proposed planning pipeline through simulations using the Autonomous
Vehicle Simulator module and through real-world experiments using autonomous trucks. Simulated experimental
results showed the following main results. First, the proposed planning pipeline can generate connected multi-level
road networks and find routes faster when using the top level of the road network. Second, the Off-Road Path Planner
can find proper paths to reach the start of the route and/or the desired final goal pose. Third, the Frénet Frames Path
Planner computed safe and comfortable alternative paths, and, in the presence of static obstacles it was chosen an
alternative path that obeyed the safe distance restrictions imposed by the autonomous vehicle system. The proposed
planning pipeline has demonstrated its effectiveness in enabling a successful mission from the current vehicle pose to
the desired final goal pose, indicating its significant value to the domain of autonomous vehicles.
