Summer School on Artificial Intelligence and Games 2018

I have assisted to the first international summer school on Artificial Intelligence and Games, which has taken place in Chania (Crete, Greece) between May 28 and June 1. During this week, I have learnt a lot of things about artificial intelligence applied to games for creating artificial agents, generating scenarios and studying human players. Additionally, we took part in the game jam of the summer school by developing "A Game of Drones", a drone race game where the human players has to do their best to get the goals before the artificial intelligence, which learned to play by using reinforcement learning. 

Mi tesis en 4 minutos

I have participated in the contest "Tu tesis en 4 minutos" (your PhD thesis in 4 minutes) organized by the Technical University of Madrid (UPM) in the context of "Symposium Cuéntanos tu tesis" (Tell us your PhD thesis symposium). This time I did not win one of the prizes, but I could know the works of my colleagues from other disciplines at the university and explain them my own work. And I think my presentation was not as bad as you would expect... :P

Cátedra UPM-CESEDEN Award to the best scientific paper

Our paper "Multi-robot interfaces and operator situational awareness: study of the impact of immersion and prediction" has been awarded with the Cátedra Ingeniero Remón y Zarco del Valle prize to the best scientific paper in defense and security. This prize is awarded annually by Universidad Politécnica de Madrid (UPM), which is my alma mater, and Centro Superior de Estudios de Defensa Nacional (CESEDEN), which is a center of the Spanish Army that deals with defense research and culture. 

The commitee has decided to award our paper because it presents a novel technology that will have an impact on the future of defense and security. This paper was the result of months of teamwork, starting from the multi-robot missions performed in Luxembourg in 2016 and finishing at the experiments with operators developed in Spain in 2017. Therefore, I would like to say "thanks" to my colleagues who participated in this work and "congratulations" to Elena because she contributed as much as me to this success.

Here you can see the paper:

J.J. Roldán, E. Peña-Tapia, A. Martín-Barrio, M.A. Olivares-Méndez, J. del Cerro and A. Barrientos. “Multi-Robot Interfaces and Operator Situational Awareness: Study of the Impact of Immersion and Prediction”. Sensors, 17(8), 1720. 2017. Impact Factor (2016): 2.677. Article

Simposio de Innovación Docente en Automática

I have attended to the Simposio de Innovación Docente en Automática (Symposium of Teaching Innovation in Automation) organized by the Comité Español de Automática (Spanish Automation Committee) in León (Spain) the January 10-12, 2018. In this event, I have seen the latest trends about inverted classrooms and remote laboratories, I have presented our work about a training system for industrial operators in assembly tasks, and I have visited the cathedral of León and a wind generator factory!

E. Crespo, J.J. Roldán, A. Martín-Barrio, J. del Cerro and A. Barrientos. “Virtual reality and process mining applied to operator training in complex assembly tasks”. IV Seminario de Innovación Docente en Automática (CEA-IFAC). León, January 10-12, 2018. Article     

One of the proposals of Industry 4.0 is the integration of machines and operators through network connections and information management. One of the challenges that can be addressed following this approach is the management of knowledge in the industry or, in other words, the transmission of knowledge from the expert operators to the new ones. This work has been developed a system that combines Virtual Reality and Process Mining to allow this knowledge transmission in the particular case of assembly tasks. Virtual Reality allows to create work environments that reproduce the real ones and interact with them without detracting resources of other processes or facing the risks of working with real systems. Process Mining allows to acquire the knowledge of experts, store it in models and then transmit it to novices. The developed system has been tested by means of several examples of assemblies with Lego.

This research line is under development and there will be news in the next months!

Robot 2017: Third Iberian Robotics Conference

I have presented two works in the ROBOT 2017 conference that has taken place in Seville (Spain) the November 22-24, 2017. 

The first work was performed by a BSc student under my supervision. In this work, we present two original game theoretic algorithms for task allocation in multi-robot systems. The first algorithm is competitive, defines games between each robot and its neighbors for the closest tasks, and searches the best Nash equilibrium. The second algorithm is cooperative, establishes a voting system in which the robots can vote their preferences to perform every task, and searches a task allocation with social welfare. The test in a simplified scenario show that both algorithms work under limited communications and partial information, but the competitive one generates better allocations than the cooperative one.

K. Garapati, J.J. Roldán, M. Garzón, J. del Cerro and A. Barrientos. “A Game of Drones: Game Theoretic Approaches for Multi-Robot Task Allocation in Security Missions”. Robot 2017: Third Iberian Robotics Conference. Seville, November 22-24th 2017. Springer Cham, p. 855-866, 2017. Article

The second work was performed by two students and I just gave them advice about technologies and algorithms. This work explores the 3D reconstruction of real scenarios and their integration in virtual reality interfaces. Two sensors are used for 3D reconstructions: laser scanners and RGBD cameras. Several algorithms are applied to convert the raw pointclouds into useful meshes.  The resultant 3D maps are visualized in virtual reality by means of a head-mounted display.

F. Navarro, J. Fdez, M. Garzón, J.J. Roldán and A. Barrientos. “Integrating 3D reconstruction and virtual reality: a new approach for immersive teleoperation”. Robot 2017: Third Iberian Robotics Conference. Seville, November 22-24th 2017. Springer Cham, 2017. Article

My short but intense career as technology entrepreneur

During the last months I have attended to the Technology Marketing Course organized by the UPM_innovatech program, where I have learnt about technology marketing, sales, negotiation, economic valuation... Additionally, I have taken part in the UPM_innovatech 2T Challenge with GreenBot, a team of robots that run through the greenhouses ensuring that the environmental conditions are appropriate for the crops, reaching the final of the contest. 

Durante los últimos meses he asistido al Curso de Comercialización de Tecnologías organizado por el programa UPM_innovatech, donde he aprendido sobre marketing tecnológico, ventas, negociación, valoración económica... Además, he participado en la competición UPM_innovatech 2T Challenge con mi idea GreenBot, un equipo de robots que recorren los invernaderos asegurando que las condiciones son óptimas para los cultivos, alcanzando la final del concurso.

SwarmCity: monitoring future cities with intelligent flying swarms

Ladies and gentlemen, let me present you... SwarmCity Project!

Future cities will be bigger and their management more complex, posing challenges related to the traffic, management of resources, maintenance of green zones, pollution, etc. The aim of this project is to provide a tool to collect the relevant information of these cities to allow their authorities to make correct and efficient decisions. The proposal goes further than the current solutions based on fixed sensors, making use of an aerial robot swarm to obtain information at desired locations and times. The work is focused on the development of swarming algorithms to solve complex tasks through the combination of simple individual behaviors, as well as data mining techniques and immersive interfaces for processing and visualizing the information. Moreover, the scenario will be used as a testbed for new algorithms designed to create an emergent and distributed intelligence that will learn from the environment.

In other words, SwarmCity is the sum of a smart city, a robot swarm and a control interface:

• The smart city has been developed by using Unity game engine. It is a simulator of a scaled city that includes traffic, pedestrians, garbage, climate and pollution. It is used not only as a view tool but also as a data source.
• The robot swarm has to cover the city collecting data about the traffic jams, people crowds, contaminant emissions... For this purpose, we are going to study behavior-based architectures, multiple techniques of optimization, game theoretical decision making...
• The control interface must allow the operators to monitor the state of the city, as well as to configure the swarm. For this purpose, we are going to explore data mining techniques to discover information, machine learning algorithms to adapt it to operator and immersive technologies to show it in a easy to understand way.

More info in SwarmCity website!