The CONDOR project addresses the development of a framework for the cognitive cities ecosystem. Cognitive cities are characterised by constant automation of everyday urban processes and continuous learning of citizens' behaviour and habits so that they can adapt to changes in the environment and thus to new needs. The city learns by collecting data provided by its sensor networks, its geographic information systems, its municipal employees and its citizens through the various technologies available. Both the citizens and the city benefit from this constant interaction as well as the continuous learning process and are constantly evolving, which also increases the collective intelligence in the city.
A cognitive urban scenario or city takes into account a number of interrelated smart systems and other smaller cognitive systems and benefits from learning theories such as connectivism. The city learns and adapts its behaviour based on past experience and is able to perceive, understand and respond to changes in its environment. It uses the collective intelligence of the city by continuously learning from the constant interaction between agents (citizens acting as sensors) via the so-called intelligence amplification loop. Therefore, not only sensor data are taken into account, but also the behavioural, spatial and political information that defines a city on a global scale. Finally, the cognitive city adapts to environmental changes, achieving sustainability, accessibility and resilience.
To design contextual, secure, private and interactive systems in order to enable cognitive oriented environments, which can be scaled from urban/dense urban areas to future smart regions. The system follows a holistic design approach, from the physical layer and its limiting factors in terms of interference level, heterogeneous link conditions and user requirements (QoS/QoE), to the fusion of data for the final provision of services.
Electromagnetic analysis in terms of volumetric wireless channel characterization and field level estimations by means of the use of hybrid computational techniques. Analysis of spatial distributions of energy consumption and temporal dispersion characteristics will be obtained for potential transceivers deployable in the scenarios under study, in terms of communication as well as EM energy harvesting capabilities, considering variable link types, transceiver allocation and user/application dynamics.
Data harvesting cognitive (and secure) from design. Construction and preservation of contextual semantics from source and by default. Cognitive data harvesting and fusion allow a precise and complete perception of things and events, allowing their comprehension and then allowing the extraction and management of knowledge.
Improve information harvesting, exchange and fusion for the development of collective intelligence. Knowledge is achieved through collective awareness, perception and intelligence, which will involve a major effort of weighting and balancing the sources of information and their weights.
Distributed learning capabilities regulated by a common ontology. Components are deployed distributed, acting each one as a learning agent which senses and interacts with the city and the citizens. In order to complete a distributed learning process, an ontology will direct the learning steps so that the complete process remains coherent along the time.
To develop sensor networks with heterogeneous processing capabilities to be deployed inside and outside the targeted cognitive environment. This involves the design of improved communications protocol to enhance streaming of these sensors, regardless of their nature.
Design of new methods for the prediction of individual and collective behaviours, as well as detection of anomalous patterns to be warned to managers and users of the infrastructure in an explainable and interpretable manner.
Provide a framework for the secure interconnection of devices, networks, information systems and intelligent agents within the context of complex urban cognitive scenarios or cognitive cities. This will consider low computational cost cryptographic tools, countermeasures against state-of-art attacks, and moving towards human-centered security.
Ensure that the collection, process and analysis of data are addressed from the privacy-protection perspective and according to regulations (GPDR). On the one hand, specific privacy enhancing technologies will be applied depending on the kind of data; on the other, privacy-preserving data mining and process mining will be used whenever necessary. Finally, citizens will be aware of their rights with respect to privacy protection.
Raise awareness of security, privacy and ethics aspects of cognitive cities, in a user-friendly manner and at a global scope, involving stakeholders and key players.
Implementation of 3 complementary pilots that allows validation of the development in real urban environments. The capacity and usability of each one of the pilots can be tested, enabling a comparative analysis, which will provide complete and extensive information in terms of performance, usability and cost.