Affordance-based imitation learning in robots

In this paper we build an imitation learning algorithm for a humanoid robot on top of a general world model provided by learned object affordances. We consider that the robot has previously learned a task independent affordance-based model of its interaction with the world. This model is used to recognize the demonstration by another agent (a human) and infer the task to be learned. We discuss several important problems that arise in this combined framework, such as the influence of an inaccurate model in the recognition of the demonstration. We illustrate the ideas in the paper with some experimental results obtained with a real robot.

Authors

• Manuel Lopes 0001
• Francisco S. Melo
• Luis Montesano

Keywords

• Humanoid robots
• Intelligent robots
• Data mining
• Learning
• USA Councils
• Humans
• Power system modeling
• Emulation
• Bayesian methods
• Imitation Learning
• Robot Learning
• Learning Algorithms
• Humanoid Robot
• Real Robot
• Idea Of This Paper
• Dynamic Model
• Markov Chain Monte Carlo
• Transition Probabilities
• Fundamental Problem
• Graphical Model
• Object Features
• Optimal Policy
• Action Recognition
• Transition Model
• Reward Function
• Sequential Task
• Policy Learning
• User-defined Parameters
• Action Repertoire
• Inverse Reinforcement Learning
• Imitative Behavior
• External Users
• Small Ball
• Optimal Rule