Team EMA at TEDx Brasília

The goals and achievements in our Project EMA were shared globally in the first TEDx Brasília. Me and Estevão Lopes, our pilot in Cybathlon 2016, have been invited to talk about Energy, Muscle, and Self-esteem. We would like to thank the organizers to arrange such an unforgettable experience.

First tests of FES rowing system

We have partnered with Capital do Remo in order to develop our novel modality of FES-aided physical exercise for athletes with paraplegia: rowing! The video below illustrates the first demonstration, in which lower limb muscle contraction in controlled using a customized controller attached to the handle.

Robotic-aided laparoscopy using serial manipulator

Demo of robot-aided laparoscopy using the Meka A2 compliant robotic arm. The setup includes a serial link manipulator integrated with an haptic interface using ROS. The video feedback is provided by a clinical endoscope.

UnBeatables application for RoboCup 2014

Team UnBeatables was established early this year to compete at Robocup Standard Platform League (SPL). The competition involves teams of NAO robots playing soccer. This is our application video. We are thrilled to take part in the Drop-in Player competition! For more info check our Facebook page.

Closed-loop elbow joint control using FES

This is the accompanying video for the paper "Closed-loop elbow joint control using FES" presented at the XI Brazilian Symposium of Intelligent Automation (SBAI). In the paper, we have evaluated PID controllers and antagonist muscles stimulation to control elbow position.

Flexible needle steering using discrete steps

This is the video accompanying the paper entitled "On the Use of Discrete Steps in Robot-aided Flexible Needle Insertion" (EMBC 2013).


Real-time feedback using Kinect

This video was recently presented by Roberto Baptista in ICRA 2013 in a workshop on Computational Techniques in Natural Motion Analysis and Reconstruction. The video demonstrates a simple methodology to facilitate learning of specialized movements using Kinect to provide real-time feedback.

Webcam as input for virtual keyboard

This video by Carlos Gonçalves demonstrates the operation of a virtual keyboard using a regular webcam and readings of the user's head movement. The idea is to design an input system for people with impairments that cannot use conventional mouse or keyboard. Following developments include more robust techniques to detect command gestures.

Robot teleoperation using Kinect

This is the video accompanying the paper entitled "Manipulator control based on the dual quaternion framework for intuitive teleoperation using Kinect" (LARS/SBR 2012). In the paper, Kinect is used to enable robot teleoperation. Since only hand position measurements are available, the operator's left hand is used to set the robot operation mode. Current works involve integrating wireless motion sensors to enable obtaining hand orientation as well.

Interactive Manipulation Between a Human and a Humanoid: When Robots Control Human Arm

This is the video accompanying the paper entitled "Interactive Manipulation Between a Human and a Humanoid: When Robots Control Human Arm" (IROS2011).

In the paper, a novel approach in human/robot collaboration is presented, where the robot controls not only its arm but also the human’s by means of functional electrical stimulation (FES). The video illustrates the method, as well as experimental results in a “ball in the hoop” task using healthy blindfolded subjects.

Joint motion estimation with inertial sensors and its integration with Kinect

This video illustrates first trials which explored the combined use of inertial sensors and Kinect to estimate joint motion for rehabilitation applications. This integration has indeed a great potential due to the complimentary roles both sensing systems feature. In this particular video, accelerometers and gyrometers are used to estimate joint angle, while the Kinect is used for initializing the inertial system and for enabling 3D visualization of the performed task.

Towards a cooperative framework for interactive manipulation involving a human and a humanoid

This is the video accompanying the paper entitled "Towards a Cooperative Framework for Interactive Manipulation Involving a Human and a Humanoid" (ICRA2011).

In the paper, the interaction is represented by means of the relative conīŦguration between the human's and the robot's hands. Based on this principle and a set of mathematical tools also proposed in the paper, a large set of tasks can be represented intuitively. The concept of simultaneous handling using mirrored movements, where the human controls the robot and simultaneously interacts with it by means of a common manipulated object, is also introduced.

Human-robot colaboration: first tests with HOAP3

Some results obtained in a human-robot colaboration experiment. All tasks were executed using the cooperative dual task-space framework proposed by Bruno Adorno and Philippe Fraisse. The setup included a HOAP3 humanoid robot, an easyTrack 500 optical tracking system, and real-time visualization.

Tremor compensation using FES: co-contraction to increase joint stability

Demo of a tremor compensation system using electrical stimulation. Based on information from portable sensors, the controller applies higher currents for stronger tremors. Tremor is induced with an auxiliary stimulator.

Using FES to control joint dynamics: system identification

In this demo we show an experiment to estimate both passive and active parameters of the wrist joint controlled by surface electrical stimulation.

Beating heart motion prediction for robust visual tracking

In the context of motion compensation systems for applications on cardiac robotic-assisted Minimally Invasive Surgery (MIS), vision-based techniques have a great potential. In this video, an Extended Kalman Filter (EKF) is used to estimate a model for heart 3D motion in order to support the vision algorithm.

Tremor online characterization + voluntary motion filtering

Demo of proposed real-time algorithm to characterize pathological tremor, while filtering voluntary motion components. The main application of this algorithm is on active tremor compensation systems.