Image-based real-time path generation using deep neural networks

• Ano de defesa: 2022
• Autor(a) principal: Moraes, Gabriel Andrade Nunes de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• 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: Veículos autônomos; Redes neurais; subject.br-rjbn; Ciência da Computação
• Link de acesso: http://repositorio.ufes.br/handle/10/16030

We propose an image-based real-time path planner for the self-driving car Intelligent Autono mous Robotic Automobile (IARA), named DeepPath. DeepPath uses a convolutional neural network (CNN) for inferring paths from images. During the self-driving car operation, Deep Path receives an image and the current car pose. Then, it sends the image to a CNN trained to infer a model of the path. After that, DeepPath generates the path in the IARA’s coordinate system using the path model. Subsequently, given the current IARA’s pose, DeepPath trans forms each pose of the path in the IARA’s coordinate system into another pose in the world coordinate system. Finally, it sends the path to the IARA’s Behavior Selector subsystem, the next subsystem in the IARA’s Decision-Making system. We evaluated the performance of DeepPath in real world scenarios. Our results showed that DeepPath is able to correctly generate paths for IARA that differ only slightly from those defined by humans.