Thesis Code: 24021
Research Area: Connected Systems and Cybersecurity

Motivation
Sensor fusion is a crucial aspect of robotics, especially in autonomous navigation. Sensors are inherently affected by noise, errors, and typically measure only specific aspects of the environment. To enhance their output and obtain a comprehensive perception of both the surroundings and the robot’s status, data from multiple sensors can be collected and fused. This process optimizes measurements, also ensuring redundancy and robustness. Sensor fusion is particularly useful for integrating localization sensors to compute both robot’s position and orientation on a map. Recently, various algorithms for sensor fusion have been developed.

Objectives
The aim of this thesis is to explore the development, integration, and testing of factor graphs for sensor fusion, one of the most widely used techniques in current literature. This algorithm will be employed to fuse data from sensors such as wheel odometry, GNSS-RTK, cameras, IMU, and UWB-based localization, in order to localize an autonomous rover during navigation. The objective is to investigate existing solutions in the literature, integrate them into RO2-based environment, and evaluate their performance.
The main steps include:
• Exploration of state-of-the-art sensor fusion techniques, with a focus on factor graphs.
• Investigation of open-source libraries that facilitate the implementation of factor graphs in a ROS2-based environment.
• Development and integration of these libraries into the middleware software developed by LINKS.
• Testing the solution in a simulated environment (e.g., Gazebo) and, potentially, on a prototype rover such as the TurtleBot or Scout 2.0.

Requirements
– Computer Science, Mechatronics, or similar background
– C++/Python programming languages
– ROS/ROS2 knowledge will be considered a plus
– Linux OS knowledge will be considered a plus
– Some knowledge in communication protocols and sensor fusion will be considered a plus
– Proactive mindset, problem-solving oriented

Duration: 6-8 months.

Contacts: please send a resume with attached the list of exams to enrico.ferrera@linksfoundation.com and gaia.zinni@linksfoundation.com