Grounding of Language in Sensorimotor World Interaction of a Humanoid Robot, using Neural Networks
نویسنده
چکیده
Abstract Symbol systems provide techniques to emulate processes of the human mind, such as abstract reasoning, natural language and problem solving, in artificial intelligent agents. To purposefully employ these techniques in artificial agents, however, these symbols need to be grounded, i.e. the agent must be able to connect a symbol to its real world referent and vice versa. Marocco et al. (2010) [11] proposed a method that uses neural networks to ground symbols in the sensorimotoric experience of a humanoid robot. After training, the networks were capable to identify three different situations and utter a corresponding proto-word. We tried to employ the proposed method on a different humanoid robot, NAO. In the process, we identified and investigated the parameters that influence the stability of the proposed method and the quality of its results.Symbol systems provide techniques to emulate processes of the human mind, such as abstract reasoning, natural language and problem solving, in artificial intelligent agents. To purposefully employ these techniques in artificial agents, however, these symbols need to be grounded, i.e. the agent must be able to connect a symbol to its real world referent and vice versa. Marocco et al. (2010) [11] proposed a method that uses neural networks to ground symbols in the sensorimotoric experience of a humanoid robot. After training, the networks were capable to identify three different situations and utter a corresponding proto-word. We tried to employ the proposed method on a different humanoid robot, NAO. In the process, we identified and investigated the parameters that influence the stability of the proposed method and the quality of its results. Zusammenfassung Mit Hilfe von Symbolsystemen können Prozesse des menschlichen Geistes, wie abstrakte Betrachtung, natürliche Sprache oder strukturierte Problemlösung, von künstlichen intelligenten Agenten nachgeahmt werden. Jedoch erfordert eine sinn volle Implementation dieser Techniken, dass die verwendeten Symbole hinreichend fundiert werden, was bedeutet, dass der Agent in der Lage sein muss, die Verbindung zwischen den abstrakten Symbolen und realen Entitäten herzustellen und nachzuvollziehen. Marocco et al. (2010) [11] haben eine Methode vorgestellt, mit der man unter Verwendung von neuronalen Netzen eine solche Fundierung von Symbolen in der sensorimotorischen Wahrnehmung eines humanoiden Roboters durchführen kann. Diese Netze waren nach dem Training in der Lage, drei Situationen voneinander zu unterscheiden und mit der Ausgabe eines entsprechenden Proto-Wortes zu reagieren. Wir haben versucht, diese Methode auf einem anderen Roboter, NAO, anzuwenden, und gleichzeitig untersucht, welche Parameter maßgeblichen Einfluss auf die Stabilität des vorgestellten Mechanismus’ und die Qualiät der Resultate haben.
منابع مشابه
Grounding Action Words in the Sensorimotor Interaction with the World: Experiments with a Simulated iCub Humanoid Robot
This paper presents a cognitive robotics model for the study of the embodied representation of action words. The present research will present how an iCub humanoid robot can learn the meaning of action words (i.e. words that represent dynamical events that happen in time) by physically interacting with the environment and linking the effects of its own actions with the behavior observed on the ...
متن کاملLanguage Emergence and Grounding in Sensorimotor Agents and Robots
The grounding of linguistic symbols in the organism’ cognitive system, and indirectly in the physical and social environment in which individuals live, is one of the most important issues in recent experimental and computational approaches to language. This is normally referred as the Symbol Grounding Problem (Harnad 1990). In cognitive science, psychological experiments have focused on the rel...
متن کاملSensorimotor Input as a Language Generalisation Tool: A Neurorobotics Model for Generation and Generalisation of Noun-Verb Combinations with Sensorimotor Inputs
The paper presents a neurorobotics cognitive model to explain the understanding and generalisation of nouns and verbs combinations when a vocal command consisting of a verb-noun sentence is provided to a humanoid robot. This generalisation process is done via the grounding process: different objects are being interacted, and associated, with different motor behaviours, following a learning appr...
متن کاملGrounding language in action and perception: from cognitive agents to humanoid robots.
In this review we concentrate on a grounded approach to the modeling of cognition through the methodologies of cognitive agents and developmental robotics. This work will focus on the modeling of the evolutionary and developmental acquisition of linguistic capabilities based on the principles of symbol grounding. We review cognitive agent and developmental robotics models of the grounding of la...
متن کاملA Method for Acquiring Body Movement Verbs for a Humanoid Robot through Physical Interaction with Humans
In this paper, we propose a language grounding method that relates verbs which imply body manipulation to motor angle patterns in a humanoid robot to make robots more entertaining. In established methods, verbs are represented by statistical models based on trajectories or motor patterns of a trajector. In our method we use a novel representation model that has six features including both traje...
متن کامل