The University of Málaga (UMA) is a public institution with a vocation to serve society. It offers a broad and higher education in a range of technical, artistic and humanistic fields. Presently the UMA has about 40.000 students and offers more than 50 diploma and degree courses and more than 40 doctorate programs. Moreover, the UMA every year imparts a wide variety of postgraduate courses in addition to the official ones.

The UMA has an active and important role in the range of international exchange and co-operation programmes, involving universities from all around the world, both in teaching and research, with special emphasis in universities from Europe and South America countries.

The Departamento de Tecnología Electrónica (DTE, http://www.dte.uma.es) is one of the greatest departments of the UMA. It is composed of 45 professors, teaches in areas such as Information and Communication Technologies and Production Technologies. The DTE is active in research areas as Multimedia and Broadband Communications, Mobile Communications, Advances Man-Machine Interfaces, Robotic and Autonomous Systems, VLSI design, Artificial Neural Networks and Evolutionary Algorithms.

Inside the DTE, the Robotic and Autonomous Systems (RAS) group is composed of 15 professors, 9 of them Ph.D. and 6 working toward their Ph.D. In addition, our group is presently directing more than 20 students doing their final project studies.

Our research focuses on two major fields: artificial vision and autonomous navigation. In the field of artificial vision, our group has developed several new retinal-topologies for fast multi-resolution image processing. Hardware implementations have been carried out using FPGAs and VLSI standard cells to realise the transformation from a homogenous resolution image into a multi-resolution one. In order to work with non-uniform geometries, we also have developed a 3D data structure based on linked pyramids. Using this structure, we have also worked in hierarchical image segmentation based on colour, texture and motion. Our segmentation algorithms rely either on adaptations of the adaptive stabilisation principle or on adaptations of the split and merge approach to our incomplete pyramids. After an image is segmented, we use different criteria to extract objects of interest from the background. These objects can be characterised by means of their corners and shapes. We have developed several fast corner detection algorithms and curvature characterisation methods that are greatly resistant against noise. Using object descriptors, we have also worked on object recognition both in 2D and 3D. The 2D recognition algorithms rely on extracting a short feature vector from the contour of a given object by projection onto a basis that represents the subspace of closed contours. Resulting vectors are classified using a variation of the k-means algorithm. 3D object recognition combines these vectors through different observations using Hidden Markov Models to accumulate information about the nature of an object from a sequence of views. Our goal in artificial vision is to develop fast and reliable algorithms that can be used on a robot.

In the field of robotics, we have basically worked with sonar sensors in navigation, environment model construction, path planning, reactive behaviours and localisation. We have developed a distributed layered hybrid architecture to combine all our algorithms into a single operating scenario. The environment is modelled by means of a grounded topological map extracted from an evidence grid in a fast and simple way. In order to operate with grids, the robot must be localised. This is locally achieved by means of Kalman filters, but we have also developed a global localisation system based on sonar landmarks and Markov fields. We have also created hierarchical path planning algorithms to operate with the available topological-metrical model of the environment in a fast way. Low level navigation and tracking is performed by means of a reactive layer. Initially, this layer relied on the potential fields approach, but we have developed a new one using Case Based Reasoning. The whole system is currently operating correctly in indoor dynamic environments, totally, partially or non explored at all. We are recently extending this approach to outdoor environments using GPS.

We have recently acquired a humanoid robot and a robotic arm. Consequently, we plan to use our developments both in vision and autonomous behaviour to make them operate in dynamic environments.

Our team has been funded from the Europe Union (Human Capital and Mobility, FEDER Operative Program), from the Spanish and Regional Andalusian Governments through several research projects, and from the private industry through technology transference projects.

Key people

Professor Francisco Sandoval was born in Spain in 1947. He received the title of Telecommunication Engineering and Ph.D. degree from the Technical University of Madrid, Spain, in 1972 and 1980, respectively. From 1972 to 1989 he was engaged in teaching and research in the fields of opto-electronics and integrated circuits in the Universidad Politécnica de Madrid (UPM) as an Assistant Professor and a Lecturer, successively. In 1990 he joined the University of Málaga as Full Professor in the Department of Tecnología Electrónica where he is Head of Department from 1991. In 1988 he was visiting professor at the University of Sheffield (UK). He is currently involved in autonomous systems and foveal vision, and maintain interest in soft computing (Artificial Neural Networks and Genetic Algorithm), and in Broad Band and Multimedia Communications. Professor Sandoval has organized more than ten International Workshops and Conferences in the domain of Artificial Neural Networks and Microelectronic Design and has been invited speaker in several Spanish Universities. He is author or co-author of over 65 scientific journal papers and more than 110 communications in International Conferences. He is co-editor of the book From Natural to Artificial Neural Computation (LNCS 930, Springer 1995). Dr. Sandoval has been main researcher in several European Union and Spanish Government funded research projects.

Dr. Cristina Urdiales was born in Spain in 1971. She received her title of Telecommunication Engineering from the Technical University of Madrid, Spain, in 1995, and her Ph.D. degree from the University of Málaga, Spain, in 1999. In 1996 she joined the Department of Tecnología Electrónica of the University of Málaga where she becomes as an Assistant Professor and a Lecturer, successively. Her research is focused on robotics and artificial vision. Dr. Urdiales is author or co-author of over 15 scientific journal papers and more than 20 communications in International Conferences.

Dr. Antonio Bandera was born in Spain in 1971. She received her title of Telecommunication Engineering and Ph.D. degree from the University of Málaga, Spain, in 1995 and 2000, respectively. During 1996 he worked in a research project under a grant by the Spanish Ministerio de Educación y Cultura. From 1997 to the present day he has worked as an Assistant Professor and Lecturer, successively, in the Department of Tecnología Electrónica of the University of Málaga. His research is focused on robotics and artificial vision, where he is author or co-author of several scientific journal papers and communications in International Conferences.