Current students

The agricoltural field is a very hard work environment, where the operator invests a lot of time and energy in order ro produce more with less. Considerable benefits can be gained in this field by automating certain operations, replacing or simply supporting the human operator. Thus borns the concept of precision agriculture that provides solutions to assist farmers in improving efficiency and reducing labor costs while respecting the legal requirements. An interesting application of this concepts are the operations in orchards, in particular vineyards. 
Automating any orchard operation requires the development of robust and precise localization and navigation algorithms. It's been proven that the robustness of the localization algorithm is not guaranteed by means of only one of the commounly used sensors (GPS,LiDAR,Cameras...), this means that an intelligent fusion is need in order to build a a robust localization system.
Navigation in orchards is very challenging due to the highly variating and hardely modeling environment. Harvesting, pruning, falling branches, missing trees or simply variations due to seasonal changes, cause not uniform conditions in terms of recurrent features. In addition, there is a high variation of the light conditions within rows, related to different coverage of trees and kind of canopy.
A lot of research is beeing made is this field in the recent years but it doesn't exist a state of the art for an orchard fully autonomous tractor. 
In my research I'm proposing an indepth analysis of the main problems in localization and navigation for autonomous tractors and propose a robust localization and navigation algorithms.