Abstract

Continuous learning of object affordances in a cognitive robot is a challenging problem, the solution to which arguably requires a developmental approach. In this paper, we describe scenarios where robotic systems interact with household objects by pushing them using robot arms while observing the scene with cameras, and which must incrementally learn, without external supervision, both the effect classes that emerge from these interactions as well as a discriminative model for predicting them from object properties. We formalize the scenario as a multi-view learning problem where data co-occur over two separate data views over time, and we present an online learning framework that uses a self-supervised form of learning vector quantization to build the discriminative model. In various experiments, we demonstrate the effectiveness of this approach in comparison with related supervised methods using data from experiments performed using two different robotic platforms.

Highlights

  • One of the fundamental enabling mechanisms of human and animal intelligence- and one of the great challenges of modern-day robotics- is the ability to perceive and to exploit environmental affordances [13]

  • We describe scenarios where robotic systems interact with household objects by pushing them using robot arms while observing the scene with cameras, and which must incre‐ mentally learn, without external supervision, both the effect classes that emerge from these interactions as well as a discriminative model for predicting them from object properties

  • We formalize the scenario as a multi-view learning problem where data co-occur over two separate data views over time, and we present an online learning framework that uses a self-supervised form of learning vector quantization to build the discriminative model

Read more

Summary

Introduction

One of the fundamental enabling mechanisms of human and animal intelligence- and one of the great challenges of modern-day robotics- is the ability to perceive and to exploit environmental affordances [13]. To recognize how to interact with objects in the world, that is to recognize what types of interactions they afford, is tantamount to understanding cause and effect relationships; from what we know of human and animal cognition, practice and experience help forge a path towards such understanding. Through countless hours of motor babbling, children gain a wealth of experience from basic interactions with the world around them, from which they are able to learn basic affordances and, gradually, more complex ones This is indicative of a continuous learning process involv‐ ing the assimilation of novel concepts over time, though it is not yet evident precisely how this learning process proceeds. This is an inherently multidisciplinary challenge, drawing on such fields as computer vision, machine learning, artificial intelligence, psycholo‐ gy, neuroscience, and others

Objectives
Methods
Results
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call