Institutional Repository
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/E6A78686-AC24-4F57-A8E6-A133E29DF694 | ||
| Year | 2023 | ||
| Type of Item | Diploma Work | ||
| License |
|
||
| Bibliographic Citation | Alexandros Theocharous, "Autonomous navigation of an electric urban car", Diploma Work, School of Production Engineering and Management, Technical University of Crete, Chania, Greece, 2023 https://doi.org/10.26233/heallink.tuc.96694 | ||
| Appears in Collections |
The aim of this thesis, is to design and implement a test-bed which will allow the development of autonomous agents, with the ability to control and navigate urban electric vehicles, in real world scenarios, using a diverse set of on-board sensors. A simulated model of a commercial urban electric vehicle, acquired by the Intelligent Systems and Robotics Laboratory of the Technical University of Crete, has been developed and the possibility of transforming it to an autonomous vehicle has been investigated as a test case. The Carla simulated environment has been used as a tool, for developing realistic simulations and simultaneously asses the functionality of different sensors, that may be integrated to the real vehicle. To demonstrate the applicability of our approach, a gym environment has been created and integrated with the suggested configuration and based on this, we’ve used the data acquired by the sensors, to train an agent for the vehicle, that allows it to perform autonomous driving tasks at a basic level. Our long term vision is to create a detailed digital twin where the simulated agent will interact with its physical counterpart and will allow the rapid prototyping of algorithms which will add autonomous capabilities to the real system.
Copyright © DIAS 2013
