Advanced Techniques for Computational Neuroethology

The project studies the application of deep learning architectures, sparse modeling techniques, lattice computing, signal processing, and machine learning to the fusion of video-image and other signals, such as depth sensors, pressure sensors and electrophysiological neural activity signals in order to obtain more precise behavior description (ethograms) and to extract correlation and causality analysis results. Work founded by MINECO under project number TIN2017-85827-P. Active from September 2017 to September 2021.

Selected Publications

M. Graña, M. Aguilar-Moreno, J. de Lope, I. Baglietto, and X. Garmendia. Improved activity recognition combining inertial motion sensors and electroencephalogram signals. International Journal of Neural Systems, 30(10):2050053, 2020. [ DOI ]

P. Romero, J. de Lope, and M. Graña. Deep learning prediction of gait based inertial measurements. In J. M. Ferrández, J. R. Álvarez Sánchez, F. de la Paz, J. Toledo, and H. Adeli, editors, Understanding the Brain Function and Emotions, LNCS 11486, pages 284--290. Springer Nature, Cham, 2019. [ DOI ]

I. Baglietto, X. Garmendia, J. de Lope, and M. Graña. Fusion of inertial motion sensors and electroecephalogram for activity detection. In J. M. Ferrández, J. R. Álvarez Sánchez, F. de la Paz, J. Toledo, and H. Adeli, editors, Understanding the Brain Function and Emotions, LNCS 11486, pages 319--326. Springer Nature, Cham, 2019. [ DOI ]

M. Graña and J. de Lope. A short review of some aspects of computational neuroethology. In J. M. Ferrández, J. R. Álvarez Sánchez, F. de la Paz, J. Toledo, and H. Adeli, editors, Understanding the Brain Function and Emotions, LNCS 11486, pages 275--283. Springer Nature, Cham, 2019. [ DOI ]

Cyber-Physical Systems for Pedagogical Rehabilitation in Special Education

The aim of the CybSPEED project is to advance a novel framework for analysis, modelling, synthesis and implementation of Cyber-Physical Systems for pedagogical rehabilitation in special education. It emphasizes the approach to learning by designing human-robot situations (games, pedagogical cases, artistic performances) and advanced interfaces (brain-computer, eye-gaze tracking, virtual reality). Partners: UPV/EHU (Spain), ISER-BAS (Bulgaria), Kyutech (Japan), Universidad de Chile, and others. Work founded by EU under project number H2020-MSCA-RISE-2017. Active from September 2017 to September 2020.

Selected Publications

J. de Lope and M. Graña. Behavioral activity recognition based on gaze ethograms. International Journal of Neural Systems, 30(7):2050025, 2020. [ DOI ]

J. de Lope and M. Graña. Comparison of labeling methods for behavioral activity classification based on gaze ethograms. In E. A. de la Cal, J. R. Villar, H. Quintián, and E. Corchado, editors, Hybrid Artificial Intelligent Systems, LNAI 12344, pages 132--144. Springer Nature, Cham, 2020. [ DOI ]

M. Graña, M. Woźniak, S. Ríos, and J. de Lope. Computational intelligence in remote sensing: An editorial. Sensors, 20(3):633, 2020. [ DOI ]

S. Moraleda, J. de Lope, and M. Graña. Recognizing cognitive activities through eye tracking. In J. M. Ferrández, J. R. Álvarez Sánchez, F. de la Paz, J. Toledo, and H. Adeli, editors, Understanding the Brain Function and Emotions, LNCS 11486, pages 291--300. Springer Nature, Cham, 2019. [ DOI ]

Unmanned Autonomous Vehicles and Drones

It could be considered as an extension of our previous project on helicopter controllers but here all the sensory system is on board. We also utilize a different family of controllers, which are mainly guided by cameras, and also new navigation strategies. Active from September 2013 to September 2017.

Selected Publications

D. Maravall, J. de Lope, and J. P. Fuentes. Navigation and self-semantic location of drones in indoor environments by combining the visual bug algorithm and entropy-based vision. Frontiers in Neurobotics, 11:46, 2017. [ DOI ]

D. Maravall, J. de Lope, and J. P. Fuentes. Vision-based anticipatory controller for the autonomous navigation of an UAV using artificial neural networks. Neurocomputing, 151(1):101--107, 2015. [ DOI ]

D. Maravall, J. de Lope, and J. P. Fuentes. Visual bug algorithm for simultaneous robot homing and obstacle avoidance using visual topological maps in an unmanned ground vehicle. In J. M. Ferrández, J. R. Álvarez Sánchez, F. de la Paz, J. Toledo, and H. Adeli, editors, Bioinspired Computation in Artificial Systems, LNCS 9108, pages 301--310. Springer, Cham, 2015. [ DOI ]

J. P. Fuentes, D. Maravall, and J. de Lope. Entropy-based search combined with a dual feedforward-feedback controller for landmark search and detection for the navigation of a UAV using visual topological maps. In M. Armada, A. Sanfeliu, and M. Ferre, editors, Advances in Intelligent Systems and Computing, LNCS 9108, pages 65--76. Springer, Cham, 2014. [ DOI ]

D. Maravall, J. de Lope, and J. P. Fuentes. A vision-based dual anticipatory/reactive control architecture for indoor navigation of an unmanned aerial vehicle using visual topological maps. In J. M. Ferrández, J. R. Álvarez Sánchez, F. de la Paz, J. Toledo, and H. Adeli, editors, Natural and Artificial Computation in Engineering and Medical Applications, LNCS 7931, pages 66--72. Springer, Berlin Heidelberg, 2013. [ DOI ]

D. Maravall, J. de Lope, and J. P. Fuentes. Fusion of probabilistic knowledge-based classification rules and learning automata for automatic recognition of digital images. Pattern Recognition Letters, 34(14):1719--1724, 2013. [ DOI ]

Language Games

Inspired by original ideas of Ludwig Wittgenstein, we study if a common lexicon can emerge in teams of robots. As a spin-off of our works on cooperative robotic systems, we try to verify the importance of language and communication in teams coordination. We could induce the language but why not let robots create their own one. Active from September 2010 to September 2015.

Selected Publications

D. Maravall, J. M. Mingo, and J. de Lope. Alignment in vision-based syntactic language games for teams of robots using stochastic regular grammars and reinforcement learning: The autonomous case and the human supervised case. Robotics and Autonomous Systems, 63(2):180--186, 2015. [ DOI ]

D. Maravall, J. de Lope, and R. Domínguez. Coordination of communication in robot teams by reinforcement learning. Robotics and Autonomous Systems, 61(7):661--666, 2013. [ DOI ]

D. Maravall, J. de Lope, and R. Domínguez. Self-emergence of a common lexicon by evolution in teams of autonomous agents. Neurocomputing, 75(1):106--114, 2012. [ DOI ]

J. M. Mingo, R. Aler, D. Maravall, and J. de Lope. Static and dynamic multi-robot coverage with grammatical evolution guided by reinforcement and semantic rules. In R. Magdalena, M. Martínez, M. Martínez, J. Vila, and P. Escandell, editors, Intelligent Data Analysis for Real-Life Applications, pages 336--365. IGI Global, Hershey Pennsylvania, 2012. [ DOI ]

D. Maravall, J. de Lope, and R. Domínguez. Self-emergence of lexicon consensus in a population of autonomous agents by means of evolutionary strategies. In E. Corchado, M. Graña, and A. Manhaes, editors, Hybrid Artificial Intelligent Systems, LNCS 6077, pages 77--84. Springer, Berlin Heidelberg, 2010. [ DOI ]

Modular and Cooperative Robotic Systems

This project has as its direct and immediate objective the development of a basic architecture and the design of the structure (including kinematics and dynamics) and control systems of a system of modular and cooperative robots, as well as the analysis of mechanisms for monitoring, navigation and coordination of the system. The project is developed in collaboration with other research groups from UDC and UPV/EHU, having each group its own and particular objectives. Work founded by MICINN under project number DPI2006-15346-C03-03. Active from October 2006 to September 2015.

Selected Publications

J. de Lope, D. Maravall, and Y. Quiñonez. Self-organizing techniques to improve the decentralized multi-task distribution in multi-robot systems. Neurocomputing, 163:47--55, 2015. [ DOI ]

J. de Lope, D. Maravall, and Y. Quiñonez. Response threshold models and stochastic learning automata for self-coordination of heterogeneous multi-task distribution in multi-robot systems. Robotics and Autonomous Systems, 61(7):714--720, 2013. [ DOI ]

M. Martín-Ortiz, J. de Lope, and F. de la Paz. Auction based method for graphic-like maps inspection by multi-robot system in simulated and real environments. Robotics and Autonomous Systems, 61(7):676--681, 2013. [ DOI ]

Y. Quiñonez, D. Maravall, and J. de Lope. Application of self-organizing techniques for the distribution of heterogeneous multi-tasks in multi-robot systems. In IEEE 9th Int. Conf. on Electronics, Robotics and Automotive Mechanics (CERMA), pages 66--71, Cuernavaca (Mexico), November 2012. [ DOI ]

Y. Quiñonez, D. Maravall, and J. de Lope. Stochastic learning automata for self-coordination in heterogeneous multi-tasks selection in multi-robot systems. In I. Batyrshin and G. Sidorov, editors, Advances in Artificial Intelligence, LNCS 7094, pages 443--453. Springer, Berlin Heidelberg, 2011. [ DOI ]

Y. Quiñonez, J. Baca, J. de Lope, M. Ferre, and R. Aracil. Self-alignment approach based on cooperative behaviors for the docking process of modular mobile robots. In IEEE Int. Conf. on Electronics, Robotics and Automotive Mechanics (CERMA), pages 445--450, Cuernavaca (Mexico), September 2010. [ DOI ]

R. J. Duro, M. Graña, and J. de Lope. On the potential contributions of hybrid intelligent approaches to Multicomponent Robotic System development. Information Sciences, 180(14):2635--2648, 2010. [ DOI ]

J. A. Martín, J. de Lope, and D. Maravall. Adaptation, anticipation and rationality in natural and artificial systems: Computational paradigms mimicking nature. Natural Computing, 8(4):757--775, 2009. [ DOI ]

J. A. Martín, J. de Lope, and M. Santos. A method to learn the inverse kinematics of multi-link robots by evolving neuro-controllers. Neurocomputing, 72(13--15):2806--2814, 2009. [ DOI ]

J. A. Martín and J. de Lope. A model for the dynamic coordination of multiple competing goals. Journal of Experimental & Theoretical Artificial Intelligence, 21(2):123--136, 2009. [ DOI ]

D. Maravall, J. de Lope, and J. A. Martín. Hybridizing evolutionary computation and reinforcement learning for the design of almost universal controllers for autonomous robots. Neurocomputing, 72(4--6):887--894, 2009. [ DOI ]

D. Maravall and J. de Lope. Neuro granular networks with self-learning stochastic connections: Fusion of neuro granular networks and learning automata theory. In M. Köppen, N. Kasabov, and G. Coghill, editors, Advances in Neuro-Information Processing, LNCS 5506, pages 1029--1036. Springer, Berlin Heidelberg, 2009. [ DOI ]

J. Pereda, J. de Lope, and M. V. Rodellar. Evolutionary controllers for snake robots basic movements. In E. Corchado, J. Corchado, and A. Abraham, editors, Innovations in Hybrid Intelligent Systems., volume 44 of Advances in Soft Computing, pages 167--174. Springer, Berlin Heidelberg, 2008. [ DOI ]

J. A. Becerra, F. Bellas, R. J. Duro, and J. de Lope. Snake-like behaviors using macroevolutionary algorithms and modulation based architectures. In D. Ruan, P. D'hondt, P. F. Fantoni, M. D. Cock, M. Nachtegael, and E. E. Kerre, editors, Applied Artificial Intelligence, pages 725--730. World Scientific, Singapore, 2006. [ DOI ]

Helicopter Flight Controller

We design a computer-based transmitter to flight conventional radiocontrolled helicopters. The sensory system uses an external camera to detect and track the helicopter to estimate its coordinates. One of the most important modules tries to learn the flight dynamics directly from video sequences. The project is developed in collaboration with Microbótica. Active from January 2006 to March 2009.

Selected Publications

J. A. Martín and J. de Lope. Learning autonomous helicopter flight with evolutionary reinforcement learning. In R. Moreno-Díaz, F. Pichler, and A. Quesada, editors, Computer Aided Systems Theory. EUROCAST 2009 Revised Selected Papers, LNCS 5717, pages 75--82. Springer, Berlin Heidelberg, 2009. [ DOI ]

J. de Lope, J. J. S. Martín, and J. A. Martín. Helicopter flight dynamics using soft-computing models. In R. Moreno-Díaz, F. Pichler, and A. Quesada, editors, Computer Aided Systems Theory. EUROCAST 2007 Revised Selected Papers, LNCS 4739, pages 621--628. Springer, Berlin Heidelberg, 2007. [ DOI ]

Studies on humanoid and biped robot locomotion

The goal is to define novel methods for humanoid and bidep robot locomotion. We have used artificial neural networks in order to solve the inverse kinematics of these robots. The networks are implemented in hardware by using commercial Altera FPGA chips. An initial robot prototype and the electronic control system have been also built. The project is developed in collaboration with Microbótica. Active from January 2003 to December 2006.

Selected Publications

J. Pereda, J. de Lope, and D. Maravall. Comparative analysis of artificial neural network training methods for inverse kinematics learning. In R. Marín, E. Onaindía, A. Bugarín, and J. Santos, editors, Current Topics in Artificial Intelligence. CAEPIA 2005 Revised Selected Papers, LNAI 4177, pages 171--179. Springer, Berlin Heidelberg, 2006. [ DOI ]

B. Prieto, J. de Lope, and D. Maravall. Reconfigurable hardware implementation of neural networks for humanoid locomotion. In J. Mira and J. R. Álvarez, editors, Artificial Intelligence and Knowledge Engineering Applications: A Bioinspired Approach, LNCS 3562, pages 395--404. Springer, Berlin Heidelberg, 2005. [ DOI ]

J. de Lope and D. Maravall. A biomimetic approach for the stability of biped robots. In M. A. Armada and P. González, editors, Climbing and Walking Robots, pages 593--600. Springer, Berlin Heidelberg, 2005. [ DOI ]

T. Zarraonandia, J. de Lope, and D. Maravall. Definition of postural schemes for humanoid robots. In R. Conejo, M. Urretavizcaya, and J. L. Pérez de la Cruz, editors, Current Topics in Artificial Intelligence. CAEPIA 2003 Revised Selected Papers, LNAI 3040, pages 241--250. Springer, Berlin Heidelberg, 2004. [ DOI ]

J. de Lope, R. González, T. Zarraonandia, and D. Maravall. Inverse kinematics for humanoid robots using artificial neural networks. In R. Moreno-Díaz, F. Pichler, and A. Quesada, editors, Computer Aided Systems Theory. EUROCAST 2003 Revised Selected Papers, LNCS 2809, pages 448--459. Springer, Berlin Heidelberg, 2003. [ DOI ]

J. de Lope, T. Zarraonandia, R. González, and D. Maravall. Solving the inverse kinematics in humanoid robots: A neural approach. In J. Mira and J. R. Álvarez, editors, Artificial Neural Nets Problem Solving Methods, LNCS 2687, pages 177--184. Springer, Berlin Heidelberg, 2003. [ DOI ]

Driving Aids with External Cameras and Range Sensors

Several methods for controlling car-like robots have been designed and evaluated, including a set of controllers for parking maneuvers with obstacles. Fuzzy logic and evolutionary algorithms are used for the development of the controllers which are able to drive a car from a intial position and orientation to a desired position and orientations with obstacles and even to make parallel parking maneuvers. We have used external cameras and range sensors in order to detect the vehicle and obstacles locations.

The project has been developed in collaboration with other research groups CSIC-IAI, DEPECA-UAH and UEX, having each group its own objectives. Some of these works are integrated into the Autopia program. Work founded by MEC under project number DPI2002-04064-C05-05. Active from October 2002 to September 2005.

Selected Publications

D. Maravall and J. de Lope. Multi-objective dynamic optimization with genetic algorithms for automatic parking. Soft Computing, 11(3):249--257, 2007. [ DOI ]

J. de Lope and D. Maravall. Multi-objective dynamic optimization for automatic parallel parking. In R. Moreno-Díaz, F. Pichler, and A. Quesada, editors, Computer Aided Systems Theory. EUROCAST 2005 Revised Selected Papers, LNCS 3643, pages 513--518. Springer, Berlin Heidelberg, 2005. [ DOI ]

D. Maravall, J. de Lope, and M. A. Patricio. Competitive goal coordination in automatic parking. In G. R. Raidl et al., editors, Applications of Evolutionary Computing, LNCS 3005, pages 537--548. Springer, Berlin Heidelberg, 2004. [ DOI ]

D. Maravall and J. de Lope. Automatic parking with collision avoidance. In D. Ruan, P. D'hondt, M. D. Cock, M. Nachtegael, and E. E. Kerre, editors, Applied Computational Intelligence, pages 463--470, World Scientific, Singapore, 2004. [ DOI ]

SIAMO - Integral System for Assisted Mobility

The project deals with automated wheelchairs. The aim is to develop a semi-autonomous system for wheelchairs. The system assists to the wheelchair user in the navigation tasks independently of his or her disability. Several input devices and navigation methods are proposed. The project is developed in collaboration with DEPECA-UAH. Work founded by CICYT under project number TER96-1957-C03-02. Active from October 1996 to September 1999.

Selected Publications

M. Mazo et al. An integral system for assisted mobility (automated wheelchair). IEEE Robotics & Automation Magazine, 8(1):46--56, 2001. [ DOI ]

J. de Lope and D. Maravall. Landmark recognition for autonomous navigation using odometric information and a network of perceptrons. In J. Mira and A. Prieto, editors, Bio-Inspired Applications of Connectionism, LNCS 2085, pages 451--458. Springer, Berlin Heidelberg, 2001. [ DOI ]

D. Maravall, J. de Lope, and F. Serradilla. Combination of model-based and reactive methods in autonomous navigation. In IEEE Int. Conf. on Robotics & Automation (ICRA), volume 3, pages 2328--2333, San Francisco (USA), June 2000. [ DOI | PDF ]

J. de Lope, D. Maravall, and J. G. Zato. Topological modeling with Fuzzy Petri Nets for autonomous mobile robots. In A. P. del Pobil, J. Mira, and M. Ali, editors, Tasks and Methods in Applied Artificial Intelligence, LNAI 1416, pages 290--299. Springer, Berlin Heidelberg, 1998. [ DOI ]

Computer Chess

It was not my first professional project but I had a lot of fun. My University organized the 7th World Computer Chess Championship in 1992. Then, we started several works concerning intelligent games and computer chess. We checked out many ideas and we also got important success by proposing novel tools and heuristics. Fortunately IBM did not contact with us thus we could kept our own goals. The team was co-directed by Ernesto Guisado and myself.

Some years later, I was invited to participate in a symposium on the Spanish engineer and mathematician Leonardo Torres y Quevedo. My speech was about his chess player as a forerunner of the current chess programs.

Selected Publications

J. de Lope. En torno al ajedrecista de torres quevedo. el ajedrez en la historia de la computación. In IV Simposio Ciencia y Técnica en España de 1898 a 1945: Cabrera, Cajal, Torres Quevedo, pages 325--339, Lanzarote (Spain), July 2004. [ URL ]