polito.it
Politecnico di Torino (logo)

Autonomous Driving Systems – Extended and Cooperative Perception Strategies

Mattia Rosso

Autonomous Driving Systems – Extended and Cooperative Perception Strategies.

Rel. Stefano Di Carlo, Fotios Stavrou. Politecnico di Torino, Corso di laurea magistrale in Ingegneria Informatica (Computer Engineering), 2024

Abstract:

The integration of self-driving technologies within Cooperative Intelligent Transportation Systems (C-ITS) presents a promising avenue for revolutionizing urban mobility. This project delves into the critical aspect of data fusion, focusing on merging information from diverse sensors and vehicles to improve the performance and reliability of autonomous vehicles. Leveraging the concept of Smart Mobility, our project explores the use of Local Dynamic Map (LDM) data to address the challenges related to this task. At the core of our research the information to be merged could come from disparate sources, including Lidar, radar, cameras, and vehicle-to-vehicle communication, to provide a comprehensive and accurate representation of the surrounding environment. By exploiting the power of LDM data, which encapsulates real-time information about road conditions, traffic flow, and infrastructure updates, our self-driving system gains enhanced situational awareness and predictive capabilities, crucial for navigating complex urban landscapes. This project has been conducted in the ALTEN Labs of Sèvres (FR) building upon previous advancements in autonomous vehicle technologies developed in the Labs and it gives contribution on a well-established proprietary pipeline based on the robotic framework ROS2 and the open source driving simulator CARLA. Under the guide of the experienced Project Leader and thanks to the unvaluable support of different people who temporarily joined the team during these months, we have re-designed some modules of the pipeline just mentioned to tackle the tasks of data fusion and object tracking by means of a Bayesian Particle Filter. An experimental campaign has been accurately and rigorously structured to observe the performances of the solutions adopted and to identify the real-world situations where these systems could really play a disruptive role optimizing primarily the safety and the comfort. We aim to contextualize this project within a new era of intelligent transportation systems, characterized by enhanced safety, efficiency, and accessibility for all.

Relators: Stefano Di Carlo, Fotios Stavrou
Academic year: 2023/24
Publication type: Electronic
Number of Pages: 58
Additional Information: Tesi secretata. Fulltext non presente
Subjects:
Corso di laurea: Corso di laurea magistrale in Ingegneria Informatica (Computer Engineering)
Classe di laurea: New organization > Master science > LM-32 - COMPUTER SYSTEMS ENGINEERING
Ente in cotutela: INSTITUT EURECOM (FRANCIA)
Aziende collaboratrici: UNSPECIFIED
URI: http://webthesis.biblio.polito.it/id/eprint/31116
Modify record (reserved for operators) Modify record (reserved for operators)