Bumper(less) cars: advanced control and ROS implementation

In this work, several state-of-the-art methods for collision avoidance with robot ground vehi- cles are compared. An initial implementation of the algorithms is provided, working both in a Robotics Operating System 2 customized environment and a simulation made in a custom Python environment. First,...

ver descrição completa

Detalhes bibliográficos
Autor: Albardaner Torras, Jaume
Tipo de documento: dissertação
Data de publicação:2024
País:España
Recursos:Universitat Politècnica de Catalunya (UPC)
Repositório:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglês
OAI Identifier:oai:upcommons.upc.edu:2117/421305
Acesso em linha:https://hdl.handle.net/2117/421305
Access Level:Acceso aberto
Palavra-chave:Automobile driving
Automated vehicles--Design and construction
Automòbils--Conducció
Vehicles autònoms--Disseny i construcció
Àrees temàtiques de la UPC::Enginyeria mecànica
Descrição
Resumo:In this work, several state-of-the-art methods for collision avoidance with robot ground vehi- cles are compared. An initial implementation of the algorithms is provided, working both in a Robotics Operating System 2 customized environment and a simulation made in a custom Python environment. First, the approach taken for the provided implementation is discussed and some possible changes are listed before bringing these algorithms to the simulation setup prepared by the laboratory. Once the algorithms are brought onto the simulation and run cor- rectly, a series of tests are run for several configurations. The best performing configuration is then brought to the real setup for verification.