The problem of scene understanding by artificial autonomous agents
The problem of scene understanding by artificial autonomous agents
Author(s): Marzena Bielecka, Andrzej BieleckiSubject(s): Philosophy, Philosophy of Science
Published by: Copernicus Center Press
Keywords: epistemology; autonomous agent; scene understanding; cognition; robotic autonomy
Summary/Abstract: In this paper, the problem of scene understanding in the context of autonomous robotics is considered. The implemented solutions, based on various types of artificial intelligence systems, are described and the perspectives of their development are discussed. In particular, the problem of cognition that reflects the structure and nature of the environment in which the artificial autonomous agent operates, is studied. The specifics of scene models and algorithms for their construction and verification of their adequacy are discussed. The presentation of the existing partial solutions, including the results obtained by the authors, is the starting point of the analysis. Possibilities of new approaches are analysed, especially those that are biologically inspired. The potential fundamental limitations are studied. These problems are discussed in the context of epistemology and in the frame of cybernetics. The analysis of the relationship between philosophy and artificial intelligence in the aspect of cognitive vision systems of autonomous robots is the aim of this paper.
Journal: Zagadnienia Filozoficzne w Nauce
- Issue Year: 2025
- Issue No: 79
- Page Range: 3-19
- Page Count: 17
- Language: English
