A neural-based model predictive control to tackle steering delay of the IARA autonomous car

• Ano de defesa: 2017
• Autor(a) principal: Guidolini, Rânik
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Mestrado em Informática; Centro Tecnológico; UFES; Programa de Pós-Graduação em Informática
• Palavras-chave em Português: Controle preditivo; Veículos autônomos; Redes neurais (Computação); Controladores PID; Ciência da Computação; 004
• Link de acesso: http://repositorio.ufes.br/handle/10/9852

In this work, we propose a Neural Based Model Predictive Control (N-MPC) approach to tackle delays in the steering plant of autonomous cars. We examined the N-MPC approach as an alternative for the implementation of the Intelligent and Autonomous Robotic Automobile (IARA) steering control subsystem. For that, we compared the standard solution, based on the Proportional Integral Derivative (PID) control approach, with the N-MPC approach. The PID steering control subsystem works well in IARA for speeds of up to 25 km/h. However, above this speed, IARA’s Steering Plant delays are too high to allow proper operation with a PID approach. We tried and modeled the IARA’s Steering Plant using a neural network and employed this neural model in the N-MPC approach. The N-MPC approach outperformed the PID approach by reducing the impact of IARA’s Steering Plant delays and allowing the autonomous operation of IARA at speeds of up to 37 km/h – an increase of 48% in the maximum stable speed.