Skip to main content
Przejdź do strony domowej Komisji Europejskiej (odnośnik otworzy się w nowym oknie)
polski pl
CORDIS - Wyniki badań wspieranych przez UE
CORDIS
Zawartość zarchiwizowana w dniu 2024-05-07

Knowledge-based planning systems for autonomous agents - cognitive robotics at work

Cel



Research objectives and content
The proposed project has the goal of applying the technology of advanced deductive databases and knowledge representation systems to the field of autonomous agents, in particular mobile robots. Indeed, recent research on cognitive robotics, a logic-based approach to action planning in robots, has come up with very expressive theoretical frameworks which provide a satisfactory solution to several representational problems (e.g. the frame problem) that affected former approaches to the formalization of dynamic systems based on the use of logic, but still lack effective computational support.
On the other hand, the most powerful approaches among the state-of-the-art knowledge representation systems are based either on the use of taxonomies of classes for representing knowledge (terminological systems), or on the use of rules and nonmonotonic features (disjunctive deductive databases). The main objective of the present project is to study the realization of a knowledge-based system for action planning based on the state-of-the-art research in cognitive robotics, and obtained by extending a powerful implemented deductive database, i.e. the Disjunctive Datalog system, developed at the Information Systems Institute of the Vienna University of Technology.
In particular, extending and coupling a deductive database with a module for taxonomic representation of information appears as a promising way to tackle the representational requirements of an advanced logic-based framework for robot planning and reasoning about actions. Such an extension can in principle be realized by exploiting the recent developments both in the deductive methods for automated reasoning within terminological systems, and in the development of optimized techniques for | automated nonmonotonic reasoning.
Training content (objective, benefit and expected impact)
The Database and Expert Systems Group at the Information Systems Institute of the Vienna University of Technology is one of the world leader research groups both in theoretical studies on deductive databases and nonmonotonic reasoning, and in implementation technologies for powerful knowledge-based systems. It thus provides an ideal support for the acquisition of high-level expertise in the field of deductive database systems. Links with industry / industrial relevance (22)
Although no industry is directly involved in the present project, the problem tackled and the expected outcomes of this research are relevant from the industrial viewpoint, since the problem of providing autonomous agents with intelligent behaviours is one of the central topics of applied research in robotics and artificial intelligence.
I

Dziedzina nauki (EuroSciVoc)

Program(-y)

Temat(-y)

Zaproszenie do składania wniosków

Brak dostępnych danych

System finansowania

RGI - Research grants (individual fellowships)

Koordynator

Technische Universität Wien
Wkład UE
Brak danych
Adres
16,Paniglgasse
1040 Wien
Austria
Koszt całkowity
Brak danych