Isaac Jesus Da Silva

Centro Universitário FEI · Department of Computer Science

This paper investigates the use of Deep Reinforcement Learning (DRL) applied to the humanoid robot soccer environment, where a robot must learn from basic to complex skills while it interacts with the environment through images received by its own camera. To do so, the Dueling Double DQN algorithm is used: it receives the images from the robot's ca...

One of the challenges of Deep Learning research is to develop algorithms for mobile robotic agents that operate in uncontrolled environments, in which dynamic changes and limited processing power are common restrictions. A common solution is to develop separate vision and decision modules, so that the former is based on deep neural network architec...

Object detection techniques that achieve state-of-the-art detection accuracy employ convolutional neural networks, implemented to have lower latency in graphics processing units. Some hardware systems, such as mobile robots, operate under constrained hardware situations, but still benefit from object detection capabilities. Multiple network models...

Object detection techniques that achieve state-of-the-art detection accuracy employ convolutional neural networks, implemented to have optimal performance in graphics processing units. Some hardware systems, such as mobile robots, operate under constrained hardware situations, but still benefit from object detection capabilities. Multiple network m...

Recent advances in deep learning point towards the use of computer vision systems based on Deep Neural Networks (DNNs). However, these network architectures are optimized to be executed in specialized hardware, such as in computers with Graphics Processing Units (GPU). Such hardware is rarely available in embedded computers, for instance, those use...

This work presents a vision system for humanoid robots of the RoboCup kidsize league. The objective of this work was develop, analysis, and implement a vision system with a Deep Neural Networks (DNN), allowing a real robot to identify objects, such as ball and robots. Several network architectures were tested and the best architecture presented an...

Reinforcement Learning (RL) is a well-known technique for learning the solutions of control problems from the interactions of an agent in its domain. However, RL is known to be inefficient in problems of the real-world where the state space and the set of actions grow up fast. Recently, heuristics, case-based reasoning (CBR) and transfer learning h...

This work shows that with a DNN using only two convolutional networks layers and two fully connected network layers it is possible to classify whether there is ball in the image, this way, the DNN can be used in an Intel NUC computer. The source code and data set are available for download.

Humanoid robots use a gait pattern generator to control the servo motors during the gait preserving its dynamic balance. There are several gait generation techniques that have been developed for humanoid robots. The Darwin-OP robot uses a method to generate the gait pattern based on coupled oscillators that perform sinusoidal trajectories. However...

This paper presents a humanoid robot framework, composed of a simulator and a telemetry interface. The framework is based on the Cross Architecture, and it is developed aiming for the RoboCup Soccer Humanoid League domain. A simulator is an important tool for testing cognitive algorithms without handling issues of real robots; furthermore, a simula...

Developing a vision system combined with a decision system for a humanoid robot, capable of playing soccer in the RoboCup domain, has been proved to be a challenging task. The computational limitations imposed by a embedded computer inside the robot and special conditions, such as the use of colored objects, led teams to use techniques based on col...

O projeto tem como foco oferecer uma solução rápida e prática para medição e aquisição da rotação de motores cuja medida direta seja impraticável. Tem como principal finalidade atender necessidades industriais. Atualmente, quando se quer aquisitar a rotação de um motor, utilizam-se técnicas nas quais é preciso ter acesso ao eixo do mesmo (com o uso...

Climbing ramps is an important ability for humanoid robots: ramps exist everywhere in the world, such as in accessibility ramps and building entrances. This works proposes the use of Reinforcement Learning to learn the action policy that will make a robot walk in an upright position, in a lightly sloped terrain. The proposed architecture of our sys...

This paper presents a new 2D robot simulator based on the Cross Architecture for RoboCup Soccer Humanoid League domain. A simulator is an important tool for testing cognitive algorithms in robots without the need of handling with real robot problems, moreover, a simulator is extremely useful for allowing reproducibility of any developed algorithm,...

In order to perform a walk on a real environment, humanoid robots need to adapt themselves to the environment, as humans do. One approach to achieve this goal is to use Machine Learning techniques that allow robots to improve their behavior with time. In this paper, we propose a system that uses Reinforcement Learning to learn the action policy tha...

One of the goals of humanoid robot researchers is to develop a complete – in terms of hardware and software – artificial autonomous agent able to interact with humans and to act in the contemporary world, that is built for human beings. There has been an increasing number of humanoid robots in the last years, including Aldebaran’s NAO and Romeo, In...

Este trabalho busca o estudo do comportamento da Integração Numérica por Monte Carlo em um problema que tem como meta encontrar o volume e centro de massa de um objeto cuja função não é bem comportada ou definida. Realizando estes experimentos ficou comprovada a robustez do método que apresentou ótimos resultados, e que este método pode se estender...

O trabalho consiste na implementação de um sistema de locomoção autônoma de um robô humanoide destinado a aplicações de pesquisa e a participação na ROBOCUP (campeonato de futebol de robôs a nível mundial), atendendo a todas as regras exigidas pelo campeonato.

Este projeto contempla o desenvolvimento e a implementação de um sistema de controle em malha fechada capaz de realizar todas as funções necessárias para a navegação autônoma do robô dentro do ambiente do campeonato de futebol de robos. Para alcançar tal objetivo, está sendo utilizadas técnicas de aprendizado por reforço com heurística na locomoção...

This paper describes the design and development of a new humanoid robot named Newton, that is intended for applications in research and also to be used in the RoboCup KidSize League World Competition. Newton robot has been designed to work without any dedicated sub-controller implemented in low level hardware, often used to control the servomotors...

This paper describes a monocular vision system for humanoid robots designed and built to compete in the RoboCup Humanoid KidSize League. The proposed vision system allows the robots to track a ball, identify goals, field lines, teammates and opponents, providing information such as distances and estimated location for the robots simultaneously, usi...

Utilizando uma ponta de corrente como transdutor, coleta-se o sinal gerado pela corrente do motor e, após a devida filtragem e condicionamento deste sinal, calcula-se a rotação do motor baseando-se no tempo de chaveamento de todas suas fases. Este método foi criado tendo em vista situações onde não há acesso ao eixo do motor e não é possível utiliz...