This Doctoral Thesis present a software architecture specifically designed for autonomous vehicle, based on the experience gained during DARPA Grand Challenge '04, '05 and '07. The architecture proposed is distributed, and each component, called "service", is defined by two interfaces: the control interface, that described how the services are controlled by the supervisor; the communication interface, that describes how the services can exchange information within each other. A list of services, for a possible autonomous vehicle is proposed in details. Finally, three Artificial Vision systems, developed and installed on autonomous vehicles, are proposed: path detection with intelligent agents, path detection by image segmentation and vehicle detection by background subtraction.
Studio di un'architettura software per veicoli autonomi basati su sensori di visione
2008
Abstract
This Doctoral Thesis present a software architecture specifically designed for autonomous vehicle, based on the experience gained during DARPA Grand Challenge '04, '05 and '07. The architecture proposed is distributed, and each component, called "service", is defined by two interfaces: the control interface, that described how the services are controlled by the supervisor; the communication interface, that describes how the services can exchange information within each other. A list of services, for a possible autonomous vehicle is proposed in details. Finally, three Artificial Vision systems, developed and installed on autonomous vehicles, are proposed: path detection with intelligent agents, path detection by image segmentation and vehicle detection by background subtraction.I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/233016
URN:NBN:IT:UNIPR-233016