Bruno Silva

Dr. Bruno Silva

Universidade Federal do Rio Grande do Norte
orcid

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
18 January 2021 in Sensors
Verified
Journal Article
Open access
PDF file
Luis Ortiz-FernandezLuis Ortiz-Fernandez
Luis Ortiz-Fernandez
ECElizabeth V. Cabrera-Avila
Elizabeth V. Cabrera-Avila
Bruno SilvaBruno Silva
Bruno Silva
Luiz GoncalvesLuiz Goncalves
Luiz Goncalves
Artificial marker mapping is a useful tool for fast camera localization estimation with a certain degree of accuracy in large indoor and outdoor environments. Nonetheless, the level of accuracy can still be enhanced to allow the creation of applications such as the new Visual Odometry and SLAM datasets, low-cost systems for robot detection and tracking, and pose estimation. In this work, we propose to improve the accuracy of map construction using artificial markers (mapping method) and camera localization within this map (localization method) by introducing a new type of artificial marker that we call the smart marker. A smart marker consists of a square fiducial planar marker and a pose measurement system (PMS) unit. With a set of smart markers distributed throughout the environment, the proposed mapping method estimates the markers’ poses from a set of calibrated images and orientation/distance measurements gathered from the PMS unit. After this, the proposed localization method can localize a monocular camera with the correct scale, directly benefiting from the improved accuracy of the mapping method. We conducted several experiments to evaluate the accuracy of the proposed methods. The results show that our approach decreases the Relative Positioning Error (RPE) by 85% in the mapping stage and Absolute Trajectory Error (ATE) by 50% for the camera localization stage in comparison with the state-of-the-art methods present in the literature.

16 September 2018 in Sensors
Verified
Journal Article
Open access
PDF file
ECElizabeth V. Cabrera
Elizabeth V. Cabrera
Luis Ortiz-FernandezLuis Ortiz-Fernandez
Luis Ortiz-Fernandez
Bruno SilvaBruno Silva
Bruno Silva
ECEsteban Clua
Esteban Clua
Luiz GoncalvesLuiz Goncalves
Luiz Goncalves
We propose a versatile method for estimating the RMS error of depth data provided by generic 3D sensors with the capability of generating RGB and depth (D) data of the scene, i.e., the ones based on techniques such as structured light, time of flight and stereo. A common checkerboard is used, the corners are detected and two point clouds are created, one with the real coordinates of the pattern corners and one with the corner coordinates given by the device. After a registration of these two clouds, the RMS error is computed. Then, using curve fittings methods, an equation is obtained that generalizes the RMS error as a function of the distance between the sensor and the checkerboard pattern. The depth errors estimated by our method are compared to those estimated by state-of-the-art approaches, validating its accuracy and utility. This method can be used to rapidly estimate the quality of RGB-D sensors, facilitating robotics applications as SLAM and object recognition.

22 March 2022 in IEEE Access
Verified
Journal Article
Open access
ALAlysson Nascimento De Lucena
Alysson Nascimento De Lucena
Bruno SilvaBruno Silva
Bruno Silva
Luiz GoncalvesLuiz Goncalves
Luiz Goncalves
We propose a new double-hybrid concept and architectural design of a tailsitter unmanned aerial vehicle with vertical takeoff and landing capability. Basically, it consists of a modified flying wing with a single combustion powertrain set and a multirotor with 2 powertrain sets with electric motors. With the electric propellers fixed at the leading wing edge, the tailsitter has two standard surfaces for elevation control and two vertical stabilizers that are used to give the necessary direction on vertical takeoff and landing. We have designed, built, and tested a prototype based on this new double-hybrid concept, which spends less energy on vertical taking off and landing and also on horizontal flight, thus maximizing flight endurance and travelled distance. We performed several experimental tests starting with the aircraft on the ground in vertical positioning. These tests include executing vertical takeoffs and landing, transitions from vertical to horizontal flight modes and transitions back (from horizontal to vertical), and hovering, which were carried out successfully. Transitions fourth and back from combustion to multirotor are inherent to some of those flight mode transitions, which have also been performed smoothly. We also performed tests (in bench) to estimate the maximum flight duration, which has demonstrated about 32 minutes of endurance. To this end, the proposed and currently built double-hybrid prototype has proven to be functional as an effective hybrid UAV system.


28 March 2024 in Journal of Intelligent & Robotic Systems
Verified
Journal Article
Open access
PDF file
ECElizabeth Viviana Cabrera-Ávila
Elizabeth Viviana Cabrera-Ávila
Bruno SilvaBruno Silva
Bruno Silva
Luiz GoncalvesLuiz Goncalves
Luiz Goncalves
Visual odometry (VO) is an important problem studied in robotics and computer vision in which the relative camera motion is computed through visual information. In this work, we propose to reduce the error accumulation of a dual stereo VO system (4 cameras) computing 6 degrees of freedom poses by fusing two independent stereo odometry with a nonlinear optimization. Our approach computes two stereo odometries employing the LIBVISO2 algorithm and later merge them by using image correspondences between the stereo pairs and minimizing the reprojection error with graph-based bundle adjustment. Experiments carried out on the KITTI odometry datasets show that our method computes more accurate estimates (measured as the Relative Positioning Error) in comparison to the traditional stereo odometry (stereo bundle adjustment). In addition, the proposed method has a similar or better odometry accuracy compared to ORB-SLAM2 and UCOSLAM algorithms.

13 January 2024 in Journal of Intelligent & Robotic Systems
Verified
Journal Article
Open access
PDF file
Bruno SilvaBruno Silva
Bruno Silva
ETEduardo Todt
Eduardo Todt
TNTiago Nascimento
Tiago Nascimento
CCCarla da C. F. Curvelo
Carla da C. F. Curvelo
Luiz GoncalvesLuiz Goncalves
Luiz Goncalves
The integration of robotics research with Artificial Intelligence (AI) and Computer Vision (CV) has enabled the solution of practical tasks previously challenging with traditional techniques. This interdisciplinary approach focuses on developing intelligent computational systems for decision-making, planning, and object recognition, minimizing human intervention in various domains like e-commerce, healthcare, industry, and more. South American researchers contribute significantly, with emphasis on emerging AI technologies such as Deep Learning, statistical methods, and computer vision applied to robotics. The works within this collection highlight the increasing prominence of robotics in education, with it becoming a part of basic education, and discuss a topical collection showcasing research presented at Latin American robotics events in 2021. The papers cover diverse topics, including altitude control for hybrid aerial-underwater vehicles, robust localization in challenging environments, cross-domain localization for underwater robots, image-based navigation for agricultural rovers, continuous training of educators in Educational Robotics, and the role of robotic simulators in education. The collection aims to disseminate cutting-edge robotics research and foster collaboration in the field.