Full text
| |
| Author(s): |
Rodrigo Calvo
Total Authors: 1
|
| Document type: | Doctoral Thesis |
| Press: | São Carlos. |
| Institution: | Universidade de São Paulo (USP). Instituto de Ciências Matemáticas e de Computação (ICMC/SB) |
| Defense date: | 2012-05-31 |
| Examining board members: |
Roseli Aparecida Francelin Romero;
Luiz Chaimowicz;
Zhao Liang;
Edson Prestes e Silva Junior;
Fernando José von Zuben
|
| Advisor: | Roseli Aparecida Francelin Romero |
| Abstract | |
The application of systems of multiple robots is desirable in various tasks. Some of them include: exploration, mining, land mine detection, security and rescue operations. An effective strategy for coordination is crucial to achieve performance improvements. In this project, two new strategies are proposed for the coordination of multiple robot systems, applied to the tasks of exploration, surveillance and formation. They are distributed, decentralized and performed in real time. The inspiration for both of them comes from the biological mechanisms that define a social organization of collective systems. Specifically, it was considered in this thesis a modified version of the Ant Colony System. The strategies are adaptable for scenarios where the number of robots and structure of the environment change. Regarding the first strategy, the experiments consider two performance criteria: average surveillance cycles and average iterations for each patrolling interval. Simulation results confirm that the exploration and surveillance emerge from the synergy of individual behaviors of robots. The data obtained show that the coordination strategy is efficient and and suitable to perform the tasks of exploration and surveillance. Regarding the second strategy, the system presents the characteristics desirable to maintain the formation: separation, alignment and cohesion. Empirical evidence showed that the system has good dispersive ability, which promoted an increase in coverage, and that it was able to adapt to new group topologies and environment settings (AU) | |
| FAPESP's process: | 08/01582-5 - Environment exploration using multi-robot for environment surveillance |
| Grantee: | Rodrigo Calvo |
| Support Opportunities: | Scholarships in Brazil - Doctorate |