Design and simulation of vehicle controllers through genetic algorithms

Genetic Programming (GP) is a population-based evolutionary technique, which, unlike a Genetic Algorithm (GA) does not work on a fixed-length data structure, but on a variable-length structure and aims to evolve functions, models or programs, rather than finding a set of parameters. There are different histories of driver development, so different proposals of the use of PG to evolve driver structures are presented. In the case of an autonomous vehicle, the development of a steering controller is complex in the sense that it is a non-linear system, and the control actions are very limited by the maximum angle allowed by the steering wheels. This paper presents the development of an autonomous vehicle controller with Ackermann steering evolved by means of Genetic Programming.

• Autores: amelec, viloria; Lizardo Zelaya, Nelson Alberto; Varela, Noel
• Tipo de recurso: Article of journal
• Fecha de publicación: 2020
• Institución: Corporación Universidad de la Costa
• Repositorio: REDICUC - Repositorio CUC
• Idioma: eng
• OAI Identifier: oai:repositorio.cuc.edu.co:11323/7685
• Acceso en línea: https://hdl.handle.net/11323/7685; https://doi.org/10.1016/j.procs.2020.07.064; https://repositorio.cuc.edu.co/
• Palabra clave: Design; Simulation; Vehicle controllers; Genetic algorithms
• Rights: openAccess
• License: CC0 1.0 Universal