Search

Everyday life takes place in space and time, and spatial experience lies at the heart of our existence. Understanding how we conceive spatial relations, and how we solve spatio-temporal problems, is therefore key to understanding human cognition. Spatial cognition research has advanced considerably over the past decades, with major successes particularly in computational implementations of knowledge representation and reasoning methods. Still, a range of key issues continue to pose major challenges. The goal of this report is to discuss the various options for the formalisation, implementation and automated solution of spatial problems including the following issues: the identification and specification of relevant concepts as expressed in human language; modules for automated understanding of domain descriptions; the use of spatial structures and affordances for direct spatial problem solving; and, the development of efficient planning systems capable of providing feasible solutions to spatial problems. This report documents the program and the outcomes of Dagstuhl Seminar 21492 "Representing and Solving Spatial Problems"., DagRep, Volume 11, Issue 11, pages 1-55

This special issue proposes a space for researchers to discuss the challenges of bridging pattern recognition and cognitive assistants. This bridge was travelled in both sides. From one side, how pattern recognition methods can help intelligent systems to become more cognitive. And from the other side, how cognitive models can improve pattern recognition methods so that the final system can better assist users. So, this special issue presents research work on these topics, aiming to observe their interrelations in order to create theoretical approaches, methodologies and computational tools to advance work on cognitive assistants and pattern recognition., Z. Falomir thanks the Cognitive Qualitative Descriptions and Applications (CogQDA) project funded by the University of Bremen and the YERUN Research Mobility Award (Young European Research UNiversities, second edition, 2018/2019). E. Plaza thanks the support by DIVERSIS (CSIC Intramural 201750E064). K. Gibert thanks the support of the Intelligent Data Science and Artificial Intelligence Research Center at Universitat Politecnica de Catalunya (UPC) and the grant CO-2018-23-T-00454 funded by UPC to organize the 21st CCIA in Roses (Catalonia, Spain) in October 2019.

La present investigació explora una perspectiva innovadora en el disseny d'algoritmes capaços de classificar pintures segons el seu estil artístic. Evitant alguns inconvenients d'integració humà-màquina que presenta el disseny amb aprenentatge automàtic, ℓ-SHE és un algoritme basat en sistemes de representació lògica i conceptes qualitatius que, a més de reconèixer i classificar amb èxit l'estil artístic d'una pintura, treballa amb informació semàntica, genera explicacions del perquè dels seus resultats, i pot ser entrenat fàcilment i amb bancs de dades relativament petits. La presente investigación explora una perspectiva innovadora en el diseño de algoritmos capaces de clasificar pinturas según su estilo artístico. Evitando algunos inconvenientes de integración humano-máquina que presenta el diseño con aprendizaje automático, ℓ-SHE es un algoritmo basado en sistemas de representación lógica y conceptos cualitativos que, además de reconocer y clasificar con éxito el estilo artístico de una pintura, trabaja con información semántica, genera explicaciones del por qué de sus resultados, y puede ser entrenado fácilmente y con bancos de datos relativamente pequeños. The present research explores an innovative perspective in the design of algorithms for classifying pictorial artworks according to their art style. Avoiding some of the human-machine integration issues associated with machine learning algorithm design, ℓ-SHE is an algorithm based on logical representation and qualitative concepts which in addition to successfully recognizing and classifying the art style of paintings, it works with semantic information, generates explanations of which reasonings have been followed for the classification results, and can easily be trained using relatively small databases.

It is almost impossible today to find an economic sector or aspect of society which does not involve AI techniques in some way. This pervasive technology has become indispensible in a multitude of ways, from supporting decision making to managing digital devices such as smart sensors, mechanical arms or artificial eyes. The ability of AI to emulate intelligence in the resolution of challenging problems has placed it at the centre of problem solving in all areas of our society. This book presents contributions from CCIA 2018, the 21st International Conference of the Catalan Association for Artificial Intelligence which took place in Alt Empordà, Catalonia, Spain, on 8-10th October 2018. The book aims to provide a picture of what is being achieved and what is under development in AI today. As such, its contents represent the diversity of approaches and applications currently being researched, but it also presents invited contributions which deal with some of the challenges that will have to be faced in the decade to come. The contributions included in this book are organized under the following headings: logic, satisfiability and fuzzy sets; classifiers, networks and machine learning; data science, recommender systems and case-based reasoning; natural language and sound processing; cognitive systems and agents; and computer vision and robotics. The book also covers a number of current AI challenges and new trends like big data, spatial problem solving, ethics and AI, and how blockchain impacts AI. Providing an up-to-the-minute overview of current AI technology and research, this book will be of value to all those with an interest in the subject.

The problem of finding a suitable qualitative representation for robots to reason about activity spaces where they carry out tasks such as leading or interacting with a group of people is tackled in this paper. For that, a Qualitative Spatial model for Group Robot Interaction (QS-GRI) is proposed to define Kendon’s F-formations [Kendon, 2010] depending on: (i) the relative location of the robot with respect to other individuals involved in that interaction; (ii) the individuals’ orientation; (iii) the shared peri-personal distance; and (iv) the role of the individuals (observer, main character or interactive). An iconic representation is provided and Kendon’s formations are defined logically. The conceptual neighborhood of the evolution of Kendon formations is studied, that is, how one formation is transformed into another. These transformations can depend on the role that the robot have, and on the amount of people involved.