Unmanned Autonomous Vehicles


Being Autonomous means being able to estimate the own position and orientation in space. The capability of accomplishing this task and the precision/accuracy achievable determine the potential utilizations of these vehicles. For this reason it is of paramount importance to develop efficient algorithms to meet the requirements. The range of sensors available nowadays is large and includes also visual aids such as standard cameras or even stereocameras.

Research interest

At the Retis Laboratory there are research activities aimed to the study of these issues. Precisely some research topics consist in

  • Localization Techniques using Radio Communication

    Teams of mobile cooperative robots are ideal candidates for applications where the presence of humans is impossible or should be avoided. Knowing the positions of the robots in crucial in such scenarios. Occasional situations may allow to build an infrastructure thus making absolute positions available; however, building infrastructure is costly and it is probably unavailable in emergency scenarios. GPS may be a possible solution for outdoors; however, it may not be available in locations such as in indoor spaces and street canyons. A possible solution is to derive relative positions from local communication.

    relative localization

  • Energy Guarantees for UAVs

    Autonomous systems currently self-monitor the status of the battery in ordere to estimate the available energy and thus the remaining autonomy during a mission. However, the type of estimation they perform is too simplistic and does not take into account the dynamic,and the current physical position, and the specific task of the vehicle. A more accurate energy model is necessary to enhance safety guarantees and perform energy-aware tasks.

    energy management of an UAV

  • Coverage Path Planning algorithms

    Coverage path planning is the operation of finding a path that covers all the points of a specific area. Thanks to the recent advances of hardware technology, Unmanned Aerial Vehicles (UAVs) are starting to be used for photogrammetric sensing of large areas in several application domains, such as agriculture, rescuing, and surveillance. However, most of the research focused on finding the optimal path taking only geometrical constraints into account, without considering the peculiar features of the robot, like available energy, weight, maximum speed, sensor resolution, etc.

    coverage path planning example

  • Hardware in the loop simulation

    Given the increasingly complexity of these Autonomous System particular importance turns out to be given to the sofware. Nowadays the trend is to bring the operating systems' APIs into the embedded system such as UAVs. This approach allows to develop solutions with an user friendly support, but it also determines other issues related to safety, real-time constraints of the applications, efficiency of the code etc. In this respect the Retis Laboratory is carrying out studies on several Real Time Operating System used on UAV solutions and developing a support framework for the development of applications.


  • Agriculture robotics: UAVs sensing for bad crops prevention

    Since the past 50 years, thanks to the benefits produced by the technology, the size of the farms increasing and the number of workers per area is increasing. With the increase of the global population, and thus the food request, we are reaching the limit of cultivable fields. New technologies are needed to increase the production of food and at the same time reduce costs. UAVs are starting to be used to monitor the growth of plants allowing an early discovery of plant infections, leading to the prevention of bad crops. However due to limited autonomy, there is the need to use team of drones instead of single units. We work on planning trajectories, detection algorithms specific for agricultural tasks.



List of the publications

  1. Carmelo Di Franco and Giorgio Buttazzo, "Energy-aware Coverage Path Planning of UAVs". Proceedings of the International Conference on Autonomous Robot Systems and Competitions (ICARSC 2015), Villa Real, April 8- 12, 2015.
  2. Carmelo Di Franco, Alessandra Melani and Mauro Marinoni, "Solving Ambiguities in MDS Relative Localization".Proceedings of the 17th International Conference on Advanced Robotics (ICAR 2015). Istanbul (Turkey), 27-31 July, 2015
  3. Carmelo Di Franco, Gianluca Franchino, and Mauro Marinoni, "Data Fusion for Relative Localization of Wireless Mobile Nodes", Proceedings of the 9th International Symposium on Industrial Embedded Systems (SIES 2014), Pisa, June 18- 20, 2014.
  4. Luis Oliveira, Carmelo Di Franco, Traian E. Abrudan,and Luis Almeida, "Fusing Time-of-Flight and Received Signal Strength for Adaptive Radio-Frequency Ranging", Proceedings of the 16th International Conference on Advanced Robotics (ICAR 2013), Montevideo, Uruguay, November 25- 29, 2013.

MS Thesis

List of MS thesis

  1. Luigi Pannocchi, “Real-time Software Development for Unmanned Aerial Vehicles”, Tesi di Laurea Magistrale in Ingegneria Robotica e dell'Automazione. Relatore: Giorgio Buttazzo, Scuola Superiore Sant’Anna, September 2015
  2. Indri Muska, “Sviluppo e simulazione di algoritmi di coordinamento per velivoli autonomi”, Tesi di Laurea in Ingegneria Informatica. Relatore: Giorgio Buttazzo, Scuola Superiore Sant’Anna, Giugno 2014.
  3. Matteo Tamburini, “Framework per lo sviluppo e la simulazione di algoritmi di controllo per velivoli autonomi”, Tesi di Laurea in Ingegneria Informatica. Relatore: Giorgio Buttazzo, Scuola Superiore Sant’Anna, Giugno 2014.
  4. Agostino Polizzano, “Strategie e algoritmi di localizzazione per WSN in ambiente indoor basati su misurazioni RSSI”. Tesi di Laurea in Ingegneria Informatica. Relatore: Giorgio Buttazzo, Scuola Superiore Sant’Anna, Luglio 2013.
  5. Carmelo Di Franco, “Wireless communication services to support teams of cooperating autonomous robots”, Tesi di Laurea in Ingegneria Informatica. Relatore: Giorgio Buttazzo, Scuola Superiore Sant’Anna, Ottobre 2013.


List of past and current projects

  • AVIATOR: Autonomous VehIcles In AgriculTuRe

    AVIATOR: is an innovative project, proposed by ReTiSLab and LandLAb of the Scuola Superiore S. Anna, that aims at using fleets of drones to monitoring the growth of plants in farms and finding plant infections on time to prevent bad crops. Using a team of cooperating drones that autonomously take off and inspects the entire field could drastically reduce the costs preventing extensive use of herbicides and reducing the manual effort of the farmer.

    Aviator UAV agriculture

Current People

Giorgio Buttazzo
Giorgio Buttazzo
Full Professor
Mauro Marinoni
Mauro Marinoni
Assistant Professor
Carmelo Di Franco
Carmelo Di Franco
PhD Student
Since Nov. 2013
Alessandra Melani
PhD Student
Since Nov. 2013
Luigi Pannocchi
PhD Student
Since Nov. 2015

Past People

  • Indri Muska, MS Student
  • Matteo Tamburini, MS Student
  • Agostino Polizzano, MS Student

Available thesis or interships

  1. Algorithm for Geo-referencing Aerial Images

    Description: Drones are starting to be used for photogrammetric sensing of large areas. Tools exist to merge all the photos taken during the survey. However, they don't exploit the knowledge of the position of the drone when the photos are taken. If the camera has a built-in GPS is it possible to use specific tools for georeferenced images. However, if the camera shots at a constant period the GPS information is missing. The objective of the work is to build an algorithm that take as input the drone trajectory and a set of photos and produces as output a set of georeferenced photos. The photos then will be given as input to a commercial software or a specific tailored algorithm can be implemented to merge the photos.

    Required skills: C, Java, Python, or Matlab; Optional: OpenCV.

  2. UAVs 3d-Object recognition

    The objective of the work is to bulid an algorithm that is able to recognize 3d-objects (trees, houses, cars) from a video during an UAV flight. The motivation given by the fact that UAVs don't perceive the surrounding space. The algorithm will take as input a video of the flight and the UAV trajectory and will extrapolate obstacle informations. The obstacles are represented as simple shape inside a 3d map.

    Required skills: C, Java, OpenCV.

  3. A C library for efficient operations on a 3D Map

    Description: The aim of the project is to implement a C Library that is able to create, modify, and access a 3D map. The map will contain simple 3d objects, each of them having a relative and absolute poistion(GPS coordinate). The library must permit to add/modify/delete/search objects efficiently. A full list of the methods to implement will be provided. A graphic representation of the map is not required. Efficient data structures (octree e k-d-tree), that are widely used in services as google maps, are higly recommended.

    Required skills: C, Java;


This page contains a list of videos of Experiments, interviews, etc. *In progress*

Here's an inteview from a localt TV about the project AVIATOR.

Il Futuro in Campagna del 07-04-2015 from tvprato on Vimeo.