Control of a Mobile Inverted Pendulum Robot Using Reinforcement Learning
Abstract
This paper introduces the concept of Reinforcement Learning applied to the problem of an inverted pendulum attached to a mobile robot (also known as CartPole). In this problem, the robot must learn the best actions to maintain the pendulum in balance. The Reinforcement Learning algorithms used in this project were Q-Learning, Q-Learning with heuristics (HAQL), and Deep Q Network (DQN). The experimental results demonstrate that the application of all algorithms successfully addressed the proposed control problem, with HAQL and DQN yielding the best outcomes.
Copyright (c) 2024 REVISTA CEREUS
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
DECLARAÇÃO DE TRANSFERÊNCIA DE DIREITOS AUTORAIS
Os autores do manuscrito submetido declaram ter conhecimento que em caso de aceitação do artigo, a Revista Cereus, passa a ter todos os direitos autorais sobre o mesmo. O Artigo será de propriedade exclusiva da Revista, sendo vedada qualquer reprodução, em qualquer outra parte ou meio de divulgação, impressa ou eletrônica.