Developer of the DropS platform, a virtual robotics laboratory designed for teaching, training, and empowering students, professors, and professionals in the fields of robotics and computing. Responsible for backend development, cloud management, data modeling, robotic simulation management, 3D modeling, and team management, using tools such as Python, Sanic Framework, Flask, ROS/ROS2, C/C++, MongoDB, PostgreSQL, Kubernetes, Docker, Azure, VNC, GitHub, Confluence, ClickUp, among others.Creator of the Huawei Mindspore Playground module for DropS, a user-friendly interface for teaching and learning artificial intelligence for students. This module utilizes libraries such as MindSpore, NumPy, Pandas, and Kivy. Additionally, responsible for training and educating students and professionals, both remotely and in person, on how to use DropS as a virtual robotic simulator.

Recipient of the IC-4 Scholarship for the project titled "Computational Environments Integration with Cloud-Edge and Artificial Intelligence." This project was developed in collaboration with the University of São Paulo (USP) and the Foundation of Support to the University of São Paulo (FUSP), with funding derived from the multinational enterprise HUAWEI.

The field of Robotics, in one of its roles, aims to stimulate knowledge and learning in various scientific areas for young people, teenagers, and adults. Additionally, it facilitates an exchange of information across different knowledge domains, ultimately changing the way reality is perceived. Currently, one of the areas of interest in Robotics is the study of artificial emotions, which involves fields such as psychology and neuroscience, to unravel and apply emotions in robotic systems.In these areas, the focus is directed towards sensory and affective systems, exploring emotions, feelings, and sensors that simulate human senses. The objective of this project is to investigate the process of emotion formation in robotic systems through tactile sensing, studying the formation of a stimulus and its correlation with the external environment. It is noteworthy that some authors assert that emotions conveyed through tactile sensing are as effective as facial or vocal expressions, for example.For the study of tactile sensing, a robotic architecture will be proposed, comprising piezoelectric sensors, Raspberry Pi 3B, Raspi2Dynamixel shield, and the AX12-JavA library. This setup will allow an exploration of how tactile sensing, represented by a circuit of piezoelectric sensors simulating artificial skin, can influence the emotional behavior of a robotic system.

Development of the project: "Studies on Consciousness and its Application to Robotics"Abstract: The field of Robotics, in one of its roles, aims to stimulate knowledge and learning in various scientific areas for young people, teenagers, and adults. Additionally, it facilitates an exchange of information across different knowledge domains, ultimately changing the way reality is perceived. Fields such as Artificial Intelligence (AI) and Artificial Life (AL) encompass human knowledge, allowing for the manipulation of vast amounts of data, providing clarity and accuracy in information, and giving meaning to them. However, limitations in intelligent architectures pave the way for the field of Artificial Consciousness (AC), which seeks to instill consciousness in robotic systems.In this research domain, areas such as memory, attention, emotions, and prediction are studied, enabling an enhancement of the cognitive capacity of machines, thinking in terms of a robotic mind. The objective of this project is to introduce an approach of AC in Robotics through the study and literature review of consciousness theories and cognitive architectures. It also involves proposing an analysis and formulation of an architecture for some aspect of consciousness, with the potential implementation validated through the mirror test.

Development of the project: "Implementation of Image Processing Technique for the Kid Size Category of RoboCup with Real Validation on the Bioloid ROBOTIS Premium Platform"Abstract: The developed architecture comprises two modules, Hardware and Software. In the Hardware module, the Robotic Player must identify all elements on the field, such as the ball, field, goalposts, opposing players, and teammates. To accomplish this task, the robot utilizes the Logitech c920 camera installed in its head, along with the pan/tilt module featuring two servo motors responsible for providing the robot with a 180-degree field of view both horizontally and vertically. The Raspberry Pi, in addition to controlling the humanoid's servo motors, uses the Open Source Computer Vision Library (OpenCV) to load and process the image sent by the camera. This enables tasks such as determining the position and size of a particular object. The Raspi2Dynamixel shield was designed and developed to connect the AX-12A servo motors of the robotic player with the Raspberry Pi. A Java library named AX12-JavA has been implemented to control the AX-12A.In the Software module, two types of Artificial Neural Networks (ANNs) were implemented for object detection, such as opponents, goals, balls, and field markings. One network classifies frames from images, and the other obtains position and size data, specifically for the ball. To achieve this result, the MATLAB Toolbox was used to train the ANNs with data generated by the webcam in OpenCV. Finally, to convert the MATLAB ANN to Java, two Java libraries, named Matrix and NeuralNet, were developed. These libraries assist in handling matrices in OpenCV and constructing a Java ANN from synaptic weight data in a ".mat" file generated with the help of the MATIO library.

Development of the project: "Implementation of Image Processing Technique for the Kid Size Category of RoboCup with Real Validation on the Bioloid ROBOTIS Premium Platform"

Study of Humanoid Locomotion with Real Validation on the ROBOTIS PlatformAbstract: The intelligence of robotic systems assists in the resolution and optimization of various human processes, making these processes more intuitive in both learning and interaction, enhancing understanding of the world. The study of humanoids allows for understanding engineering as an interdisciplinary application. Several locomotion techniques were investigated in this project, including Center of Mass (CM), Zero Moment Point (ZMP), Proportional, Integral, and Derivative (PID) control, and Genetic Algorithms (GA).In addition to this theoretical study, a literature review was conducted on sixty-five humanoid robots, classifying them based on name, year, Degrees of Freedom (DOF), and locomotion algorithms. The study of humanoid locomotion was validated using the Virtual Robot Experimentation Platform (V-REP) simulation environment and also on the Bioloid ROBOTIS Premium platform.The movement of human walking was analyzed to aid in collecting data from the humanoid's motors and sensors. After data collection, discrete functions were defined in fourteen time intervals, which were used to determine the equilibrium positions during the movement of the robot's feet. The primary objective in this stage was to collect data that allows the robot's CM to remain balanced. Through virtual and real experiments, specific polynomial functions for the movement of each joint of the humanoid were created. Finally, the data was validated in both V-REP and the real platform.

Planning social projects, schedules, and goals. Achieving short, medium, and long-term results. Innovating the team by introducing new solutions and tools for implementation.