3D PERCEPTION, LEARNING-BASED DATA FUSION
cod. 1010750

Academic year 2023/24
2° year of course - First semester
Professor
Maria Letizia MARCHEGIANI
Academic discipline
Sistemi di elaborazione delle informazioni (ING-INF/05)
Field
A scelta dello studente
Type of training activity
Related/supplementary
60 hours
of face-to-face activities
6 credits
hub: PARMA
course unit
in ENGLISH

Learning objectives


At the end of the course, students will be able to:
* Identify the advantages and disadvantages in the use of different sensors and their potential combinations, depending on the application and the environmental conditions;
* Critically determine strengths, limitations and constraints of different approaches to sensor and data fusion;
* Employ and deploy techniques and methods to combine signals from several different sensing modalities;
* Extract and leverage information from unimodal and multimodal data for environment interpretation;
* Design, test, and evaluate single-sensor and multi-sensor systems for vehicle perception.

Prerequisites

- - -

Course unit content


The goal of the course is the analysis of fundamental and advanced aspects of 3D perception and data fusion towards the robust and accurate sensing and interpretation of a vehicle's environment, both from a theoretical and practical perspective.

To this end, the exploration of more classic approaches to sensor data analysis and integration will be augmented with the investigation of state-of-the-art solutions from the most recent scientific literature.

The main topics in a nutshell:
* Sensors for Mobile Autonomous Systems
* Single-Sensor 3D Perception
* Traditional Data Fusion
* Learning-based Data Fusion
* From Interpretation to Interpretability and Explainability

Full programme

- - -

Bibliography


There won’t be a reference textbook, but slides and additional material (e.g. scientific papers) will be provided during the course.

Teaching methods


The theoretical aspects will be covered during the lectures and will also include student presentations; the technical aspects will be approached via a project consisting of a series of assignments.

Assessment methods and criteria


Students will be asked to present and discuss scientific papers on the topics covered by the course and to implement and present a project.

Both paper and project presentations will take place throughout the course’s lectures.
The final grade will reflect the performance in the paper presentations/discussions as well as in the assignments.

Other information

- - -

2030 agenda goals for sustainable development

- - -

Contacts

Toll-free number

800 904 084

Student registry office

E. segreteria.ingarc@unipr.it

Quality assurance office

Education manager:
Dott.ssa Jasmine Salame Younis
T. +39 0521 906045
Office E. dia.didattica@unipr.it
Manager E. jasmine.salameyounis@unipr.it

 

President of the degree course

Prof. Massimo Bertozzi
E. massimo.bertozzi@unipr.it

Faculty advisor

Prof. Letizia Marchegiani
E. letizia.marchegiani@unipr.it

Career guidance delegate

Prof. Letizia Marchegiani
E. letizia.marchegiani@unipr.it

Tutor professor

Prof. Nicola Mimmo (UNIBO)
E. nicola.mimmo2@unibo.it
Prof. Riccardo Rovatti (UNIBO)
E. riccardo.rovatti@unibo.it

Erasmus delegates

to be determined

Quality assurance manager

Prof. Nicola Mimmo (UNIBO)
E. nicola.mimmo2@unibo.it
 

Internships

Prof. Alessandro Chini (UNIMORE)
E. alessandro.chini@unimore.it
Prof. Gaetano Bellanca (UNIFE)
E. gaetano.bellanca@unife.it
Prof.ssa Annamaria Cucinotta (UNIPR)
E. annamaria.cucinotta@unipr.it
Prof. Nicola Mimmo (UNIBO)
E. nicola.mimmo2@unibo.it
Prof. Paolo Pavan (UNIMORE)
E. paolo.pavan@unimore.it
Prof. Riccardo Rovatti (UNIBO)
E. riccardo.rovatti@unibo.it

Tutor students

to be determined