Cognitive Systems

Objectives

To develop artificial systems that can interpret data arising from real-world events and processes (mainly in the form of data-streams from sensors of all types) and in particular from visual and/or audio sources); acquire situated knowledge of their environment; act, make or suggest decisions and communicate with people on human terms, thereby supporting them in performing complex tasks.

Focus

Focus is on research into ways of endowing artificial systems with high-level cognitive capabilities, typically perception, understanding, learning, knowledge representation and deliberation, thus advancing enabling technologies for scene interpretation, natural language understanding, automated reasoning and problem-solving, robotics and automation, that are relevant for dealing with complex real-world systems. It aims at systems that develop their reasoning, planning and communication faculties through grounding in interactive and collaborative environments, which are part of, or connected to the real world.

These systems are expected to exhibit appropriate degrees of autonomy and also to learn through “social” interaction among themselves and/or with people; in a longer term perspective, research will explore models for cognitive traits such as affect, consciousness or theory of mind.

Work will build on ongoing research; it is expected to be highly interdisciplinary, drawing on appropriate fields that contribute to cognitive science and cognitive engineering: artificial intelligence, computer vision and robotics, as well as relevant branches of mathematics (e.g. dynamical systems), the bio-sciences (e.g. neuroscience) and the humanities (e.g. linguistics, philosophy). It should yield new approaches towards understanding and improving cognitive capabilities in artefacts and explore new methods of integrating these in complete artificial systems.

The investigation of viable methods living up to demanding application requirements for autonomous or semi-autonomous systems is also encouraged, preferably in industrial inspection and monitoring, complex systems control, medicine or the life sciences.

Aims

The project aims at specifying, implementing and testing a general architecture for cognitive systems which are able to solve complex problems in domains where knowledge is poorly formalised and information is incomplete.

Within the project we intend:

• to establish the mathematical foundation of cognitive information processing, and
• to develop innovative knowledge representation and reasoning methods to be applied in cognitive systems.

Major worksteps

• basic research on cognitive reasoning and knowledge representation,
• development of innovative knowledge representation and reasoning tools,
• integration of components into a Cognitive Systems Technology Toolset,
• development of ontologies, languages and reasoning methods in a domain selected for testing,
• implementation of a domain-specific prototype system, and
• system testing on sample problems.

Elementsof cognitive reasoning theory

• complex logic for combining two valued logic and plausible inference rules,
• theory for combining various forms of plausible reasoning,
• general algebraic theory of similarity underlying plausible reasoning,
• theory for constructive argumentation,
• quasi-axiomatic theory for formalising domain knowledge,
• theory of introspection and self reflection, and
• parallel architecture for cognitive reasoning.

Instruments:

IPs will be used to research the modelling and architecture of entire cognitive systems. They may also research systems-level integration of methods and tools, as well as the integration of different layers of the cognition process (e.g. combining low- and high-level cognitive functions). STREPs will primarily target specific research issues, cognitive functionalities or components which are best researched within small, flexible groupings. CAs are encouraged to promote collaboration across previously fragmented communities, with a view to forming future joint research networks. Alternatively to a CA, a well-balanced NoE combining a critical mass of interdisciplinary research would be welcome. All actions should promote pertinent aspects of community and skills building, where appropriate, with an outreach to and inclusion of industry and application service provision.

Indicative budget: IPs, NoEs: 65%; STREPs, CAs: 35%

Call information: IST Call 4