Tesi etd-01312020-113228 |
Link copiato negli appunti
Tipo di tesi
Tesi di laurea magistrale
Autore
CIRAVOLO, PIETRO
URN
etd-01312020-113228
Titolo
A Framework for Validation of a Bio-Inspired Exploration
Algorithm in Multi-Robot Scenarios
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
COMPUTER ENGINEERING
Relatori
relatore Prof.ssa Bernardeschi, Cinzia
relatore Prof. Domenici, Andrea
correlatore Dott. Palmieri, Maurizio
relatore Prof. Domenici, Andrea
correlatore Dott. Palmieri, Maurizio
Parole chiave
- bio-inspired algorithm
- co-simulation
- cyber-physical systems
- design space exploration
Data inizio appello
21/02/2020
Consultabilità
Completa
Riassunto
Cyber-Physical Systems (CPSs) are complex engineered systems which involve many different physical and software components.
Co-simulation techniques, in which sub-systems possibly modelled and simulated with different tools are coordinated by a co-simulation engine, provide a powerful instrument to test the system before it is physically available.
In this thesis, a framework for the co-simulation of a bio-inspired exploration algorithm applied to a group of robots is presented. The framework relies on the integrated tool chain for model-based design of CPSs of the INTO-CPS project. One of the advantages of the framework is that the behaviour of the algorithm can be analysed at different levels of abstraction. First, the algorithm is analysed without modelling the physics of the robots, then the kinematics of the robots is introduced in order to move towards a more realistic scenario.
Results of the co-simulations are analysed and discussed.
A simple interface that provides graphic feedback on the evolution of the co-simulation over a virtual map has been implemented as an additional component of the co-simulation framework.
The design space exploration functionality of the tool chain has been exploited in order to extract good parameters of the algorithm that can ensure high map coverage in a short time.
Co-simulation techniques, in which sub-systems possibly modelled and simulated with different tools are coordinated by a co-simulation engine, provide a powerful instrument to test the system before it is physically available.
In this thesis, a framework for the co-simulation of a bio-inspired exploration algorithm applied to a group of robots is presented. The framework relies on the integrated tool chain for model-based design of CPSs of the INTO-CPS project. One of the advantages of the framework is that the behaviour of the algorithm can be analysed at different levels of abstraction. First, the algorithm is analysed without modelling the physics of the robots, then the kinematics of the robots is introduced in order to move towards a more realistic scenario.
Results of the co-simulations are analysed and discussed.
A simple interface that provides graphic feedback on the evolution of the co-simulation over a virtual map has been implemented as an additional component of the co-simulation framework.
The design space exploration functionality of the tool chain has been exploited in order to extract good parameters of the algorithm that can ensure high map coverage in a short time.
File
Nome file | Dimensione |
---|---|
TesiPiet...avolo.pdf | 1.16 Mb |
Contatta l’autore |