On-line learning of lexical items and grammatical constructions via speech, gaze and action-based human-robot interaction

نویسندگان

  • Grégoire Pointeau
  • Maxime Petit
  • Xavier Hinaut
  • Guillaume Gibert
  • Peter Ford Dominey
چکیده

In order to be able to understand a conversation in interaction, a robot, has to first understand the language used by his interlocutor. A central aspect of language learning is adaptability. Individuals can learn new words and new grammatical structures. We have developed learning methods that allow the humanoid robot iCub to robot can learn new lexical items by interaction with the human and consolidation of its autobiographical memory. Then, based on these open class words, the robot can bootstrap the acquisition of novel grammatical structures in real-time. Finally, we demonstrate how human gaze can be monitored, and could be used in order to reduce referential ambiguity inherent in such learning conditions. These learning capabilities are demonstrated in a collection of videos. 1. Learning of ”open-class” word through Human Robot interaction, using an autobiographical-like memory In robotic systems based on human-inspired robot task learning [1], significant attention has been allocated to the mechanisms that underlie the ability to acquire knowledge from and encode the individual’s accumulated experience [2], and to use this accumulated experience to adapt to novel situations [3]. In this context we consider research on autobiographical memory (ABM) and mechanisms by which ABM can be used to generate new knowledge [4]. Figure 1: Physical interaction between the iCub and an human. Both are interacting with the ReacTable (the blue table). Behind the robot, its internal representation is displayed on a screen. The objective of the current research is to demonstrate how an autobiographical memory system, coupled with mechanisms for detecting and extracting regularities can be used to construct a progressive hierarchy of spatial, and temporal relations that provide the basis for learning and executing shared plans. The mixed-initiative dialogue between the Robot and the Agent is done as described in the Figure 1 and 2, by a simple oral interaction. RTable iCub Motor Cmd

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Developing a Semantic Similarity Judgment Test for Persian Action Verbs and Non-action Nouns in Patients With Brain Injury and Determining its Content Validity

Objective: Brain trauma evidences suggest that the two grammatical categories of noun and verb are processed in different regions of the brain due to differences in the complexity of grammatical and semantic information processing. Studies have shown that the verbs belonging to different semantic categories lead to neural activity in different areas of the brain, and action verb processing is r...

متن کامل

The Comparison of Native English and Persian Elementary School Students’ Performance on Lexical and Grammatical Collocations

The importance and howness of language learning/ acquisition has been a great concern for decades. There are many factors that play important roles in this regard. This research compared the performance of native Persian and English elementary students to see if there is any significant difference between the two groups and which type of collocation they performed better within the groups. For ...

متن کامل

Learning Grammatical Constructions from Narrated Video Events for Human-Robot Interaction

The ability to provide natural language interaction with humanoids will likely require adaptive systems that learn the mappings from speech to meaning in a context dependant manner. The objective of this research is to develop a system for miniature language learning based on a minimum of pre-wired language-specific functionality, that is compatible with observations of perceptual and language ...

متن کامل

Learning Grammatical Constructions

We describe a computational model of the acquisition of early grammatical constructions that exploits two essential features of the human language learner: significant prior knowledge of concepts and individual lexical items, and sensitivity to the statistical properties of the input data. Such principles, previously applied to lexical acquisition, are shown to be useful and necessary for learn...

متن کامل

A Q-learning Based Continuous Tuning of Fuzzy Wall Tracking

A simple easy to implement algorithm is proposed to address wall tracking task of an autonomous robot. The robot should navigate in unknown environments, find the nearest wall, and track it solely based on locally sensed data. The proposed method benefits from coupling fuzzy logic and Q-learning to meet requirements of autonomous navigations. Fuzzy if-then rules provide a reliable decision maki...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2013