Unmanned Autonomous Vehicles
Recent news you should be aware of ... * 01/10/2015: Detailed schedule will be online by the end of 2015
This is a description page for the PhD course on Unmanned Autonomous Vehicles in Air, Land and Sea. This 5 CFU course can be taken also by Master students from the Computer Engineering program, i.e., students enrolled in the Laurea Magistrale.
Course Aim & Organization
The course is aimed at providing students (with basic knowledge about robotics and control) with theoretical and practical instruments, related to modelling and identification, perception, control, path/trajectory and mission planning, required to understand this evolution, and set up unmanned autonomous vehicles in the air, land and sea domains.
The course is held by (in order of appearance)
- Luca Bascetta, Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano
- Matteo Matteucci, Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano
- Marcello Farina, Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano
- Marco Lovera, Dipartimento di Ingegneria Aerospaziale, Politecnico di Milano
- Alfredo Martins, Instituto Superior de Engenharia do Porto
The course presents the knowledge required to better understand the commonalities and specificities of unmanned autonomous vehicles design in the different domains of air, land, and sea.
To provide a common base to better understand the specificities induced by the particular domain, the course will provide basic knowledge about the general tools and common components that are involved in the design of an unmanned autonomous vehicle. After an initial introduction of the current state of the art and potential applications of unmanned autonomous vehicles, the course will introduce:
- the most common vehicle kinematic and dynamic models;
- the fundamentals on path/trajectory and mission planning;
- the fundamentals on vehicle model identification and state estimation;
- the most common sensors for vehicle localization, control, obstacle avoidance;
- fundamentals of model predictive control applied to vehicle trajectory tracking, stabilization and obstacle avoidance.
The previous topics will be complemented by a description of a complete application case study for each domain (air, marine and land) with the purpose to highlight how each of the components has been applied or it has required some specific adaptation to cope with its peculiarity. In particular, we will discuss applications in the following domain
- air: fixed wing and rotary unmanned aerial vehicles;
- land: off-road unmanned vehicles;
- sea: surface and underwater unmanned vehicles.
Current trends and future applications will be presented and discussed in a final panel with people from academia and industry.
The course schedule for this year edition foresees 36 hours of lectures from 9:00 to 18:00 (slight adjustments in time might happen according to participants needs) held in a single week at the beginning of June 2016. The room for all the lectures will be communicated with the exact dates.
Course Material & Referencies
The following is some suggested material to follow the course lectures. Slides will be published by the teachers right before the lecture.
Slides and lecture notes
- Day 1 lectures by Luca Bascetta and Matteo Matteucci
- Course Intro: UAV Course introduction, teachers presentation and evaluation criteria
- Control Architectures: deliberative vs. reactive robot control architectures, hybrid approaches
- Kinematics: Holonomic and non holonomic constraints in mechanical system, unicycle and bicycle robot kinematics
- Trajectory planning: Trajectory planning, search based algorithms, randomized algorithms, planning vs. control, the Dynamic Window Approach
- DARPA Challenge: Some example control architectures from the 2007 DARPA Urban Challengen
- Day 2 lectures by Marcello Farina and Marco Lovera
- Day 3 and 4 lectures by Alfredo Martins and Jose Almeida
- B. Siciliano, L. Sciavicco, L. Villani, G. Oriolo. Robotics: Modelling, Planning and Control. Springer
- D. Fox, S. Thrun, W. Burgard. Probabilistic Robotics. MIT Press.
- L. Magni, R. Scattolini. Advanced and Multivariable Control. Pitagora Ed.
- S. M. LaValle. Planning Algorithms. Cambridge University
PhD students will be asked to study a few scientific papers on a specific subject and prepare a 20-30 minutes presentation summarizing the content of the papers with reference to the course syllabus. The presentation will be evaluated by the course Professor in Charge for their content.
Master students will be asked to carry out a project using one of the techniques presented during the course.