About me

I am currently working as a Robotics Engineer at Aldebaran Robotics and finishing my M.Sc. at Boston University in Electrical Engineering. I also have a B.Sc. in Mechatronics Engineering at Universidad Nacional de Ingeneria in Lima, Peru.

My main interests are in the fields of Robotics, Computer Vision, Image Processing, Machine Learning, Artificial Intelligence and Computer Graphics.

For more information, you can download my resume.

Projects

This is the most recent list of all the latest projects I've been working on. If you have any questions about them, send me an email and I'll try to get back to you as soon as possible.

Autonomous Navigation with NAO

This project was completed for the EC720 class at Boston University. Here we developed a navigation system for the humanoid robot NAO using data acquired from the 2 cameras mounted on the robot to estimate the position of the ball with respect to the robot. NAO seeks to find a ball, navigate to it, and finally kick it.

You can take a look at the final report here.

The video below shows the final result.

Remote control of NAO using a Gumstix Board

This project was completed for the EC535 Embedded Systems class at Boston University. Here we introduce the use of a Gumstix board as a remote control for the NAO humanoid robot to perform 4 different tasks. Using an LCD touchscreen, we set up a GUI which sends data via Bluetooth to a server running on a PC. The PC process the data into executable commands which will send to the robot via WiFi.

You can take a look at the final report here.

The video below shows the final result.

Optical Position and Tracking sensing using the EM algorithm

This project was completed for the EC503 class at Boston University. In this project we investigated the Expectation-Maximization algorithm to estimate signal positions on a 2D detector where the presence of ambiente noise and detector dark counts makes position estimation difficult. We also expanded the case to estimate multiple static beams and moving beams. For the latter case we made use of the Kalman Filter algorithm and the Hungarian algorithm.

You can take a look at the final report here.

Virtual Shape Recognition using Leap Motion

This project was completed for the EC520 class at Boston University. We designed a system to recognize hand drawing gestures of numerical letters in the air using the Leap Motion sensor. By doing a 3D to 2D processing of the input data, this data is then sent to a recognition system using k-Nearest Neighbor method to estimate the numerical value the user is writing.

You can take a look at the final report here.

Facial Identification using Multilayer Perceptron

This project was completed for the CS585 class at Boston University. We implemented and trained a Multilayer Perceptron Neural Network (NN) which classify people based on faces. We used the ATT Labs dataset and create a training and test set which are processed for face feature extraction which will serve as our data to train our NN.

This system can also be extended to input a new set of images for new persons and learn to recognize these new people.

You can take a look at the final presentation here.

Design and modelling of a 4DOF Robotic Arm

This project was completed for my Advanced Robotics class in my senior year at college. We modelled, simulated and controlled a 4 degrees-of-freedom (DOF) robotic arm using linear, non-linear and fuzzy control. The goal of this project was to study the kinematics and dynamics of an industrial robot and implement advanced methods to control its position.

You can take a look at the final report here.

Courses

EC516 Digital Signal Processing

EC505 Stochastic Processes

EC519 Speech Signal Processing

EC520 Digital Image Processing

EC720 Digital Video Processing

CS585 Image and Video Computing (Class based on C++ and OpenCV. All of the assignments' results can be found here)

EC503 Learning from Data

EC535 Introduction to Embedded Systems