Research

A CNN Pilot for Autonomous Drone Racing

Master Thesis


Supervisor
Jose Martinez-Carranza, PhD.
Full-Time Principal Researcher B in the Computer Science Department at the Instituto Nacional de Astrofisica Optica y Electronica (INAOE) based in Puebla City, in Mexico.
Honorary Senior Research Fellow in the Computer Science Department at the University of Bristol, in the UK.

Abstract

Convolutional neural networks (CNN) and deep learning (DL) have become a popular tool for addressing all kinds of artificial intelligence challenges. The Autonomous Drone Race is a challenge that consists of developing a drone capable of defeating a human in a drone race. DL is a tool that has been included in state-of-the-art solutions to address this problem. Current work has proposed using CNN and DL to detect the gates, while other work has proposed using a CNN to obtain the flight speed and a three-dimensional reference point, these data are used by the controller to generate the corresponding control signals. It should be noted that all these approaches use a single frame as input.

Motivated by the above, this work aims to develop a CNN to obtain the control signals directly for a drone to navigate autonomously in a drone racing circuit. This implies two levels of difficulty: 1) navigating through a predefined sequence of gates; 2) navigating in an environment where the location of the gates is not known a priory. The performance tests were carried out in the Gazebo simulator, using the AR drone vehicle.


This research produced the following list of publications:

Index Journals:

  1. L. Oyuki Rojas-Perez, Jose Martinez-Carranza. "Autonomous Drone Racing with an Opponent: A First Approach". 19th Mexican International Conference on Artificial Intelligence (MICAI). Published in a Special Issue of “Computación y Sistemas” Journal. CONACYT Index. Mexico City, Mexico. October, 2020.

  2. L. Oyuki Rojas-Perez, Jose Martinez-Carranza. “DeepPilot: A CNN for Autonomous Drone Racing”. Special Issue "UAV-Based Sensing Techniques, Applications and Prospective". Sensors. August, 2020. JCR Q1. https://doi.org/10.3390/s20164524

  3. A. A. Cabrera-Ponce, L. Oyuki Rojas-Perez, J.A. Carrasco-Ochoa, J. F. Martinez-Trinidad, Jose Martinez-Carranza. “Gate Detection for Micro Aerial Vehicles using a Single Shot Detector”. IEEE Latin America Transactions. December, 2019. JCR Q4. https://doi.org/10.1109/TLA.2019.9011550

Peer-Reviewed International Conferences:

  1. L. Oyuki Rojas-Perez, Jose Martinez-Carranza. "Temporal CNN based Learning for Autonomous Drone Racing". IEEE 5th Workshop on Research, Education and Development of Unmanned Aerial Systems (RED-UAS). IEEE, Cranfield, UK. November, 2019.

  2. J. Arturo Cocoma-Ortega, L. Oyuki Rojas-Perez, Aldrich A. Cabrera-Ponce, Jose Martinez-Carranza. "Overcoming the Blind Spot in CNN-based Gate Detection for Autonomous Drone Racing". IEEE 5th Workshop on Research, Education and Development of Unmanned Aerial Systems (RED-UAS). IEEE, Cranfield, UK. November, 2019.


GitHub: https://github.com/QuetzalCpp/DeepPilot

Autonomous Navigation System for Micro Aerial Vehicles

Engineering Thesis


Supervisor
Jose Martinez-Carranza, PhD.
Full-Time Principal Researcher B in the Computer Science Department at the Instituto Nacional de Astrofisica Optica y Electronica (INAOE) based in Puebla City, in Mexico.
Honorary Senior Research Fellow in the Computer Science Department at the University of Bristol, in the UK.

Abstract
Design and implementation of an autonomous navigation system in environments without access to GPS based on visual SLAM. The proposed system involved the implementation of a location system that it allows to know the location of the MAV in meters, using only the image transmitted by the vehicle to the real-time workstation, the odometry and the height at which the MAV is located. The approach of this work is to enrich the exploration and acquisition of information with aerial images in environments without access to GPS.


This research produced the following publication:

L. Oyuki Rojas-Perez, Jose Martinez-Carranza. “Metric Monocular SLAM and Colour Segmentation for Multiple Obstacle Avoidance in Autonomous Flight”. IEEE 4th Workshop on Research, Education and Development of Unmanned Aerial Systems (RED-UAS). Linköping, Sweden. October, 2017.