This group is part of the research group GIAP, at the Department of Electrical and Electronic Engineering, Universidad de los Andes, Colombia. Our research interests are focused on theoretical and applied developments in Machine Learning, Pattern Recognition, and Control Systems.
Luis Felipe Giraldo Trujillo
lf.giraldo404 [at] uniandes [dot] edu [dot] co
We list some projects and opportunities for students at Undergraduate, Master, and PhD levels to conduct relevant research. If you are interested in getting involved in any of these projects or want to propose new ones, please, feel free to contact us at lf.giraldo404[at]uniandes[dot]edu[dot]co .
We gratefully acknowledge ongoing and past financial support from:
Machine Learning in Smart Cities
Based on concepts in multivariate statistical analysis and optimization, we develop Machine Learning and Pattern Recognition techniques to process and analyze a large amount of information (Big Data) that lead to “intelligent” decision making. Ongoing applications include:
· Photovoltaic power systems
· Artisanal pisciculture
· Precision agriculture
· Water and air pollution analysis
· Automatic sentiment and stress recognition and analysis
· Nanotechnology and materials
· Control of utility consumption patterns of residential consumers
· Smart meter signal analysis
· Crime dynamics
· Healthcare systems
· Detection of landmines in humanitarian demining
· 3D structure reconstruction in computer vision
· Distributed and secure learning of pattern recognition systems
Design of Cooperation Networks in Multi Agent Systems
There are many complex processes whose behavior can be modeled and analyzed as the result of the interaction of simpler interconnected dynamical systems. Examples include power systems, cooperative autonomous vehicles, and human communities. We develop mathematical and computational tools for the design and control of such interactions, considering aspects such as stability, structured sparsity, vulnerability to attacks, and performance.
Data-driven Control of Dynamical Systems
Many processes involve uncertainty that make their control a challenging task, especially when the control design method requires a model of the process that is known a priori. Based on concepts in Machine Learning, including Reinforcement Learning, we study and develop control strategies whose action depend on inferences made on the observed data. Ongoing projects include analysis and control of industrial processes, autonomous vehicles, and structures with active components in civil engineering.
Luis Felipe Giraldo Trujillo
Department of Electrical and Electronic Engineering
Universidad de los Andes, Colombia
Email: lf.giraldo404 at uniandes dot edu dot co
Álvaro Javier Florez (PhD student)
Gabriel Narvaez (PhD student)
Andrés Jiménez (PhD student)
Gilberto José Díaz (MSc student)
Andrés Felipe Zambrano (MSc student)
Jesús Gabriel Angel (MSc student)
Abel Felipe Zambrano (MSc student)
Maria Paula Franco
Cristian Yesid Andrade
Juan David Medina
Past graduate students
Catalina Albornoz (Now Energy Efficiency Studies Coordinator at GreenYellow, Colombia)
Gabriel Narvaez (Now PhD student at Uniandes, Colombia)
Alvaro Javier Florez (Now PhD student at Uniandes, Colombia)
María Angélica Arroyo (Now at Lockheed Martin, USA)
Anonymous comments and ratings by the students on these courses and the instructor can be seen at losestudiantes.co .
Recent conference publications and poster presentations
· AJ Florez, LF Giraldo. Model Predictive Control and Structural Sparsity. In IEEE Colombian Conference on Automatic Control, Colombia, 2019.
· CD López, LF Giraldo. Optimization of Energy and Water Consumption on Crop Irrigation using UAVs via Path Design. In IEEE Colombian Conference on Automatic Control, Colombia, 2019.
· G Díaz-García, L Burbano, N Quijano, LF Giraldo. Distributed MPC and Potential Game Controller for Consensus in Multiple Differential-Drive Robots. In IEEE Colombian Conference on Automatic Control, Colombia, 2019.
· AF Zambrano, LF Giraldo. Solar Irradiance Forecasting Models without On-Site Training Measurements. Renewable Energy, Vol. 152, pp 557-566, 2020.
· AJ Florez, LF Giraldo. Structural Sparsity in Networked Control Systems. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2018.
· JD Clapp, D Ruderman, H Gonzalez, LF Giraldo, KM Passino, M. Reed, I Fernandes. A System Dynamics Model of Drinking Events: Multi-Level Ecological Approach. Systems Research and Behavioral Science, Vol. 35, Issue 3, pp 265-281, 2018.
· H Gonzalez, LF Giraldo, KM Passino. Feedback Control Engineering for Cooperative Community Development. IEEE Control Systems Magazine. Vol. 38, Issue 3, pp 87-101, 2018.
· LF Giraldo, KM Passino, JD Clapp, D Ruderman. Dynamics of Metabolism and Decision Making During Alcohol Consumption. IEEE Transactions on Cybernetics, Vol 47, Issue 11, pp 3955-3966, 2017.
· LF Giraldo, KM Passino. Dynamics of Cooperation in a Task Completion Social Dilemma. PLoS ONE, e0170604, 2017.
· LF Giraldo, KM Passino, JD Clapp. Modeling and Analysis of Group Dynamics in Alcohol-Consumption Environments. IEEE Transactions on Cybernetics, Vol 47, Issue 1, pp 165 - 176, 2017.
· LF Giraldo, KM Passino. Dynamic Task Performance, Cohesion, and Communications in Human Groups. IEEE Transactions on Cybernetics, Vol 46, Issue 10, pp 2207 - 2219, 2016.
· LF Giraldo, F Lozano, N Quijano. Foraging Theory for Dimensionality Reduction of Clustered Data. Machine Learning, Vol. 82, Issue 1, pp 71 - 90. 2011.
· LF Giraldo, E Delgado, JC Riaño, G Castellanos. Selección de características usando modelo híbrido basado en algoritmos genéticos. Revista Ingeniería e Investigación. Vol. 26, Issue 3, pp 113 -119. 2006.
· LF Giraldo, E Delgado, G Castellanos. Cinemática Inversa de un Brazo Robot utilizando Algoritmos Genéticos. Revista Avances en Sistemas e Informática. Vol. 3, Issue 1, pp 29 - 34, 2006.
· JD Clapp, KM Passino, LF Giraldo, D Ruderman. Developing a Collaborative Systems Dynamics Model of Drinking Events: Finding a Common Language. Book chapter in Innovations in Collaborative Modeling, The University of Michigan Press. In press.