Artificial intelligence in padel sport: Creation of an AI system

• Autor(a) principal: Castro, João Rodrigo Ferreira Fernandes de
• Data de Publicação: 2023
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10071/30751
• topic: Inteligência artificial -- Artificial intelligence; Machine learning; Padel; YOLO

Padel is a sport that has been arousing everyone's curiosity and, therefore, the demand for it is high. More and more sports fields and venues have been built for the practice of this modality. As such, it is still an underexplored area in terms of Machine Learning technology and prediction models. It is possible to generate statistics in a Padel game, just like in other sports, and this dissertation proposes a strategy to develop a Machine Learning model to predict two movements relating to the game, namely Left Volleys and Trays. It is also essential to identify the trajectory of the ball in order to carry out a detailed analysis, as well as detect the lines of the padel court, in order to create an Artificial Intelligence system capable of predicting the movements made by the players and generating statistics on them. Thus, the general aim of this work is to design and develop a Machine Learning model to optimize the game of Padel, which together with the detection of the court and the identification of the ball, will generate the Artificial Intelligence system. The methodology is divided into three phases. The exploratory phase involves gathering information from the data and building data pipelines to extract insights, develop models, and test variables using feature engineering. This is followed by the AI modeling/automation phase to implement in the system. The third phase is analyzing the results to integrate AI into any real-time system.