Automated warehouse systems are becoming increasingly popular, in the last years. Along these lines, researches on the coordination of the fleet of automated guided vehicles (AGVs), as a solution for addressing the problems of production efficiency and flexibility, are becoming more and more important.
Our work mainly focuses on methodologies for modeling the traffic of fleets of AGVs. Our approach consists in considering the whole system to improve the global performance and not to focus only on a single aspect of the multi-robot system (e.g., navigation algorithms, path planning, roadmap, etc.). In fact, optimizing a local cost function is not sufficient for maximizing the global performance. The objective is then that of defining advanced coordination strategies, for obtaining high performances for the overall system.
This is achieved considering, in a holistic manner, all the aspects that mainly influence the performance of AGV systems, namely:
- Traffic modeling
We are working on the development of methodologies based on which, knowing the current position of each AGV in a warehouse, it is possible to predict, in a reliable manner, the future traffic conditions
- Task allocation
AGVs are used for solving transportation tasks, that is moving goods from one location to another. The problem is that of defining which AGV needs to perform which transportation task. We are solving this problem explicitly including the traffic model, to maximize the overall efficiency.
- Roadmap design
AGVs are constrained to move along a roadmap, that is typically manually designed by experts. We are developing strategies for automating such process.
Attività Programmazione di una postazione robotica per l’implementazione di un applicazione di pick and place tra macchina blisteratrice e astucciatrice per applicazioni biomedicali. Azienda Tutor [...]
PAN-Robots (European Project, FP7, 2012-2015)
Trafcon (Echord experiment, FP7, 2010-2011)