Internal Pages

Login to access



Revision as of 14:51, 27 April 2015 by WikiSysop (Talk | contribs)
(diff) ← Older revision | Current revision (diff) | Newer revision → (diff)

Development of Hierarchical Concepts in Humanoid Robots


Project No: 111E287
Project Duration: 3 years (01.05.2012-01.05.2015)
Project Budget: 244.980 TL
Principal Investigator: Sinan Kalkan

One of the important problems of a developing robot or an organism is the development of concepts and linking it with language. Concepts are very important for intelligence and cognition for, due to them, an organism can generalize and abstract over his experiences; he can transfer his experiences to other situations or events; he can think, reason, remember, and talk about situations, events or entities; and, maybe the most important of all, concepts are essential in language comprehension.

Although there is abundant theoretical literature on concepts, there is little computational work on how a developing robot or an organism can acquire concepts and related these acquired concepts to language and these problems, as can be understood from the lately funded EU and TUBITAK projects, are beginning to attract the attention of many research groups. The existing studies focus just on first-order concepts and mostly focus on affordance-based or haptic-based concept formation. However, first-order concepts (such as APPLE, ORANGE, DOG, CAT2 etc.) are not sufficient for a robot which we expect to interact with us, humans; an intelligent and cognitive organism should be able to abstract over abstractions (and acquire concepts like FRUIT, ANIMAL etc.), i.e., acquire concepts of the concepts. For example, when shown an apple and told “Look this is an apple”, the robot should be able to link the word “apple” to the APPLE concept; however, when later told a different sentence for an apple like “Look this is a fruit”, the robot should be able to handle the conflict in using different words for the same concept. We claim that resolving such conflicts and communication/interaction with humans at this level require hierarchical representations of concepts.

As known from developmental psychology, main modalities affecting concept formation are: language, appearance and function; i.e., babies use the symbols from adults (i.e., words), the shape and the texture, and the affordances of the objects for grouping and conceptualizing them. This way, a single object (for example, a little red rotten apple) can be a part of many different concepts (e.g., APPLE, ROLLABLE, FRUIT, THROWABLE, GROWS-ON-TREE) based on the context and the goal.

In this project, we will study how a (cognitively) developing and embodied humanoid robot can acquire a hierarchical representation of concepts from its experiences. For this goal, by going beyond the current literature, we will use language, appearance as well as affordances of objects and investigate how these three modalities can affect the formation of the hierarchy. The proposed methods and mechanisms will be demonstrated on a concise scenario involving a humanoid robot iCub interact with a human on a clearly defined task.


  • H. Celikkanat, G. Orhan, N. Pugeault, F. Guerin, E. Sahin, S. Kalkan, "Learning Context on a Humanoid Robot using Incremental Latent Dirichlet Allocation", IEEE Transactions on Autonomous Mental Development (TAMD), (accepted with revision), 2015. Available as a technical report: METU-CENG-TR-2015-01.
  • H. Celikkanat, S. Kalkan, “Integrating Spatial Concepts into a Probabilistic Concept Web”, 17th International Conference on Advanced Robotics (ICAR), (accepted), 2015.
  • H. Celikkanat, G. Orhan, S. Kalkan, "A Probabilistic Concept Web in a Humanoid Robot", IEEE Transactions on Autonomous Mental Development (TAMD), DOI: 10.1109/TAMD.2015.2418678, (in press), 2015. PreprintPublisher's Website
  • H. Celikkanat, G. Orhan, N. Pugeault, F. Guerin, E. Sahin and S. Kalkan, "Insansi Robotlarda Baglamin Ogrenilmesi", 1nci Turkiye Otonom Robotlar Konferansi (TORK), ODTU, 6-7 Kasim 2014. pdf
  • S. Kalkan, N. Dag, O. Yuruten, A. M. Borghi, E. Sahin, "Verb Concepts from Affordances", Interaction Studies Journal, 15(1):1-37, 2014. PreprintPublisher's webpage
  • H. Celikkanat, S. Kalkan, "Using Slowness Principle for Feature Selection: Relevant Feature Analysis", 22nd IEEE Conference on Signal Processing and Applications (Special Session on Cognitive Robotics and Applications), Trabzon, Turkey, 2014. pdf
  • H. Celikkanat, G. Orhan, N. Pugeault, F. Guerin, E. Sahin and S. Kalkan, "Learning and Using Context on a Humanoid Robot Using Latent Dirichlet Allocation", Int. Conference on Development and Learning and Epigenetic Robotics (ICDL-Epirob), 2014. pdf
  • O. Yuruten, E. Sahin, S. Kalkan, "The Learning of Adjectives and Nouns from Affordance and Appearance Features", Adaptive Behavior, 21(6):437-451, 2013. PreprintPublisher
  • G. Orhan, S. Olgunsoylu, E. Sahin, S. Kalkan, "Co-learning nouns and adjectives", IEEE International Conference on Development and Learning and on Epigenetic Robotics, Osaka, Japan, 2013. pdf
  • S. Kalkan, O. Yuruten, E. Sahin, "Relating Affordances with Verbs, Nouns and Adjectives", Special Session on Cognitive Robotics and Applications, IEEE 21st Conference on Signal Processing and Communication Applications, Girne, KKTC, 2013 (in Turkish). Available as pdf
  • Yuruten O., Uyanik K. F., Caliskan, Y., Bozcuoglu, A. K. Sahin E., Kalkan S., “Learning Adjectives and Nouns from Affordances on the iCub Humanoid Robot”. 12th Int. Conference on Simulation of Adaptive Behavior, Denmark, 2012. Available as: pdf.


  • Orhan, G. 2014. “Building a web of concepts on a humanoid robot”, M.Sc., Computer Engineering, Middle East Technical University.
  • Çelikkanat, H. 2015. “Grounded and Contextualized Web of Concepts on a Humanoid Robot”, Ph.D., Computer Engineering, Middle East Technical University (expected: September 2015).

Invited Talks, Seminars

Videos (coming soon)