Valeria Arruda - Academia.edu (original) (raw)

Papers by Valeria Arruda

IEEE Access

This paper addresses the control of a fleet of unmanned aerial systems (UAVs), termed as drones, ... more This paper addresses the control of a fleet of unmanned aerial systems (UAVs), termed as drones, for flight formation problems. Getting drones to fly in formation is a relevant problem to be solved when cooperative cargo transportation is desired. A general approach for this problem considers the coordination of a fleet of UAVs, by fusing all information coming from several individual sensors posed on each UAVs. However, this approach induces a high cost as every UAV should have its advanced perception system. As an alternative, this paper proposes the use of a single perception system by a fleet composed of several elementary drones (workers) with primitive low-cost sensors and a leader drone carrying a 3D perception source. We propose a Quadral-Fuzzy approach to ensure that all drones fly in formation and will not collide with each other or with environment obstacles. We also develop a new way to compute potential fields based on possibility fuzzy (fuzziness) measure with the focus of avoiding collisions between the drones. The proposed approach encompasses four high-coupled intelligent controllers that respectively control the leader and worker drones' motion and implement obstacle and collision avoidance procedures. Simulation results using a fleet of four aerial drones are presented, showing the potential for solving usual problems to flights in formation, such as dodging obstacles, avoiding collisions between the drones, among others.

Anais do 14º Simpósio Brasileiro de Automação Inteligente

This work presents a platform based on augmented reality for simulation, analysis and test of nav... more This work presents a platform based on augmented reality for simulation, analysis and test of navigation systems to multiple robots in a real lab-scale environment. The main objective of this platform is to provide conditions for the use of simple and low-cost robots, with hardware constraints, in autonomous navigation applications. The framework Robot Operating System (ROS) is used to integrate all platform's components . An application example is presented to demonstrate the available resources during experiments, where multiple real Sphero robots receive their navigation skills in a virtual and independent way, ensuring a decentralize navigation behavior. The achieved results confirm the feasibility of this experimental tool. Resumo: Este trabalho apresenta uma plataforma baseada em realidade aumentada para simulação, análise e teste de sistemas de navegação para múltiplos robôs em um ambiente real, em escala de laboratório. O principal objetivo desta plataformaé fornecer condições para que robôs simples e de baixo custo, com limitações em seu hardware, possam ser utilizados em aplicações de navegação autônoma. O framework Robot Operating System (ROS)é usado para integrar todos componentes da plataforma. Um exemplo de aplicaçãoé implementado para demonstrar os recursos disponíveis nos experimentos, em que múltiplos robôs reais do tipo Sphero recebem suas habilidades de navegação de forma virtual e independente, garantindo um comportamento descentralizado de navegação. Os resultados alcançados confirmam a factibilidade desta ferramenta experimental.

Journal of Intelligent & Robotic Systems

Proceedings XXII Congresso Brasileiro de Autom�tica

2016 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), 2016

2016 12th IEEE International Conference on Industry Applications (INDUSCON), 2016

2016 XIII Latin American Robotics Symposium and IV Brazilian Robotics Symposium (LARS/SBR), 2016

IEEE Transactions on Cognitive and Developmental Systems, 2016

Boletim De Ciencias Geodesicas, Nov 8, 2004

Abstract O multicaminho atua na fase da onda portadora causando erros na distância receptor-satél... more Abstract O multicaminho atua na fase da onda portadora causando erros na distância receptor-satélite medida e fornecendo medidas de fase alteradas. Talvez por isso, o multicaminho continue sendo a maior fonte de erro nos posicionamentos de alta precisão, ...

ABSTRACT Several applications use robotic vision such as a robot navigating through an unknown su... more ABSTRACT Several applications use robotic vision such as a robot navigating through an unknown surrounding can use vision as main navigate sensor. This paper focuses on studying camera calibration via stereo vision by means of neural network. A neurocalibration method is proposed based on the neural network ability to learn nonlinear relationship among a two and three dimension coordinate systems and also its information generalization skill. The data used to train neural network mapping are generated from a calibration grid point obtained through the use of Harris edge detection algorithm. The experimental results indicated that the neurocalibration method is feasible and efficient.

2011 Ieee International Conference on Imaging Systems and Techniques, May 1, 2011

International Journal of Systems Science, Nov 26, 2010

ABSTRACT

IEEE Latin America Transactions, 2016

Studies in Computational Intelligence, 2016