Interactive object classification using sensorimotor contingencies

Understanding and representing objects and their function is a challenging task. Objects we manipulate in our daily activities can be described and categorized in various ways according to their properties or affordances, depending also on our perception of those. In this work, we are interested in representing the knowledge acquired through interaction with objects, describing these in terms of action-effect relations, i. e. sensorimotor contingencies, rather than static shape or appearance representations. We demonstrate how a robot learns sensorimotor contingencies through pushing using a probabilistic model. We show how functional categories can be discovered and how entropy-based action selection can improve object classification.

Authors

• Virgile Hogman
• Mårten Björkman
• Danica Kragic

Keywords

• Robot sensing systems
• Shape
• Data models
• Predictive models
• Gaussian processes
• Uncertainty
• Object Classification
• Sensorimotor Contingencies
• Effect Of Activity
• Learning Process
• Posterior Probability
• Changes In Position
• Sensor Data
• Point Cloud
• Learning Objectives
• Gaussian Process
• Object Shape
• Object-oriented
• Action Classes
• Measure Of Confidence
• Highest Confidence
• Optimal Action
• Changes In Translation
• Rotation Changes
• Gaussian Process Model
• Continuous Action Space
• Rotation Characteristics
• Bayes Classifier
• Proper Activity
• Object Parts
• Similarity Matrix