A Grounded and contextualized web of concepts on a humanoid robot

Çelikkanat, Hande
In this thesis, we propose a formalization for a densely connected representation of concepts and their contexts on a humanoid robot platform. Although concepts have been studied implicitly and explicitly in numerous studies before,our study is unique in placing the relatedness of concepts to the center: We hypothesize that a concept is fully meaningful only when considered in relation to the other concepts. Thus, we propose a novel densely connected web of concepts, and show how utilizing the relatedness of concepts can take cognition one step forward from the conventional approach that treats them individually. Then we use this densely connected framework for determining the context of encountered scenes. Although unanimously accepted as one of the pillars of cognition, our study is the first attempt to provide a dedicated and general formalization of context in a robotics setting. We follow a developmental approach in which the robot determines the existing contexts in its environment in an unsupervised manner, associates seen objects and whole scenes with these contexts as appropriate, and further utilizes this extracted contextual information in reasoning and planning. As required by the developmental paradigm, the programmer’s input to the robot in terms of informational bias is kept at a minimum, and the robot is expected to deduce the important characteristics of the environment itself, such as the number of contexts hidden in its environment, if and when to introduce another context to its world model, and how these contexts probabilistically give rise to the related concepts in this world.


Building a web of concepts on a humanoid robot
Orhan, Güner; Kalkan, Sinan; Department of Computer Engineering (2014)
In this thesis, an effective approach for predicting nouns, adjectives and verbs is introduced for more effective communication between a humanoid robot and a human actor. There are three important challenges addressed by our approach: The first one is the accurate prediction of words in language. Most of the existing robotics studies predict words in language using perceptual information only. However, due to noise and ambiguity in low-level sensory information, prediction using perceptual information is o...
A developmental framework for learning affordances
Uğur, Emre; Şahin, Erol; Öztop, Erhan; Department of Computer Engineering (2010)
We propose a developmental framework that enables the robot to learn affordances through interaction with the environment in an unsupervised way and to use these affordances at different levels of robot control, ranging from reactive response to planning. Inspired from Developmental Psychology, the robot’s discovery of action possibilities is realized in two sequential phases. In the first phase, the robot that initially possesses a limited number of basic actions and reflexes discovers new behavior primiti...
Momentum transfer continuum between preshape and grasping based on fluidics
Özyer, Barış; Erkmen, İsmet; Erkmen, Aydan Müşerref; Department of Electrical and Electronics Engineering (2012)
This dissertation propose a new fluidics based framework to determine a continuum between preshaping and grasping so as to appropriately preshape a multi-fingered robot hand for creating an optimal initialization of grasp. The continuum of a hand preshape closing upon an object that creates an initial object motion tendency of the object based on the impact moment patterns generated from the fingers is presented. These motion tendencies should then be suitable for the proper initiation of the grasping task....
A Probabilistic Concept Web on a Humanoid Robot
Çelikkanat, Hande; Orhan, Guner; Kalkan, Sinan (2015-06-01)
It is now widely accepted that concepts and conceptualization are key elements towards achieving cognition on a humanoid robot. An important problem on this path is the grounded representation of individual concepts and the relationships between them. In this article, we propose a probabilistic method based on Markov Random Fields to model a concept web on a humanoid robot where individual concepts and the relations between them are captured. In this web, each individual concept is represented using a proto...
The learning and use of traversability affordance using range images on a mobile robot
Ugur, Emre; Dogar, Mehmet R.; Cakmak, Maya; Şahin, Erol (2007-04-14)
We are interested in how the concept of affordances can affect our view to autonomous robot control, and how the results obtained from autonomous robotics can be reflected back upon the discussion and studies on the concept of affordances. In this paper, we studied how a mobile robot, equipped with a 3D laser scanner, can learn to perceive the traversability affordance and use it to wander in a room filled with spheres, cylinders and boxes. The results showed that after learning, the robot can wander around...
Citation Formats
H. Çelikkanat, “A Grounded and contextualized web of concepts on a humanoid robot,” Ph.D. - Doctoral Program, Middle East Technical University, 2015.