Industrial robots have grown over the years making production systems more and more efficient, requiring the need for efficient trajectory generation algorithms that optimize and, if possible, generate collision-free trajectories without interrupting the production process. In this work is presented the use of Reinforcement Learning (RL), based on the Q-Learning algorithm, in the trajectory generation of a robotic manipulator and also a comparison of its use with and without constraints of the manipulator kinematics, in order to generate collisionfree trajectories. The results of the simulations are presented with respect to the efficiency of the algorithm and its use in trajectory generation, a comparison of the computational cost for the use of constraints is also presented.
Robotics has been reaching a lot of space in recent years and is evolving every day. Today, robotics is seen as an important area in many applications. Dynamic control is an essential tool for robotics. To design the control, a well-modeled / identified plant is required to have a well-defined controller design. A misidentified system can compromise the project, causing it to have a control effort and a very long tuning time, which is not good at all. The present work aims to show some solutions for the identification of a robotic manipulator with a hybrid system of a recursive least squares technique and a metaheuristic called differential evolution (DE). The results are compared with other methods and can surpass them in order to validate the study and showing its potential. Resumo: A robótica vem alcançando muito espaço nosúltimos anos e está evoluindo a cada dia. Atualmente, a robóticaé vista como umaárea importante em muitas aplicações. O controle dinâmicoé uma ferramenta essencial para a robótica. Para projetar o controle,é necessária uma planta bem modelada/identificada para ter um projeto de um controlador bem definido. Um sistema mal identificado pode comprometer o projeto, fazendo com que ele tenha um esforço de controle e um tempo de sintonia muito longo, o que nãoé nada bom. O presente trabalho tem como objetivo mostrar algumas soluções para uma identificação de um manipulador robótico com um sistema híbrido de uma técnica dos mínimos quadrados recursivos (MQR) e uma metaheurística denominada evolução diferencial (DE). Os resultados são comparados com outros métodos e conseguem superá-los, a fim de validar o estudo e mostrando o seu potencial.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.