Artificial intelligence applied to orbit prediction in the three-body scenario as a chaotic system.

• Grant number: 25/18178-8
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: February 01, 2026
• End date: January 31, 2027
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Luiz Otavio Murta Junior
• Grantee: Gabriel Hupfer Righi
• Host Institution: Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto (FFCLRP). Universidade de São Paulo (USP). Ribeirão Preto , SP, Brazil

Abstract

The N-Body Problem is one of the most classical and challenging problems in physics, with numerous well-documented analytical and numerical approaches. In recent years, however, machine learning techniques have gained prominence in computational physics, although their application to modeling trajectories in complex gravitational systems-particularly in three-body configurations-has not yet been thoroughly explored. Considering this gap, this project proposes the creation and training of a Deep Neural Network model capable of generalizing dynamic behaviors and producing accurate predictions of the positions, velocities, and acting forces on each body after a defined time interval, based on various initial conditions. The research is grounded, among other references, in Deep Neural Networks to Enable Real-Time Multimessenger Astrophysics (George and Huerta), The Solution of the N-body Problem (Florin Diacu), and the book Dive into Deep Learning (Aston Zhang et al.). Based on these theoretical foundations, the project aims to address the challenges of efficiently modeling chaotic gravitational systems using modern deep learning techniques, with a focus on reducing computational costs compared to traditional simulation and prediction methods. (AU)