Modeling affordances using Bayesian networks

Affordances represent the behavior of objects in terms of the robot's motor and perceptual skills. This type of knowledge plays a crucial role in developmental robotic systems, since it is at the core of many higher level skills such as imitation. In this paper, we propose a general affordance model based on Bayesian networks linking actions, object features and action effects. The network is learnt by the robot through interaction with the surrounding objects. The resulting probabilistic model is able to deal with uncertainty, redundancy and irrelevant information. We evaluate the approach using a real humanoid robot that interacts with objects.

Authors

• Luis Montesano
• Manuel Lopes 0001
• Alexandre Bernardino
• José Santos-Victor

Keywords

• Bayesian methods
• Humans
• Intelligent robots
• Humanoid robots
• Robot sensing systems
• Cognitive robotics
• Object detection
• Context modeling
• Motion measurement
• USA Councils
• Bayesian Model
• Motor Skills
• Object Features
• Perceptual Skills
• Humanoid Robot
• Behavior Of Objects
• Dimensional Space
• Set Of Equations
• Contact Information
• Markov Chain Monte Carlo
• Dimensional Vector
• Marginal Likelihood
• Bayesian Framework
• Discrete Variables
• Action Recognition
• Image Position
• Hand Position
• Statistical Dependence
• Discrete Random Variable
• Bigger Ones
• Hand Velocity
• Object Velocity
• Product Of Likelihoods