Reinforcement learning vs human programming in tetherball robot games

Reinforcement learning of motor skills is an important challenge in order to endow robots with the ability to learn a wide range of skills and solve complex tasks. However, comparing reinforcement learning against human programming is not straightforward. In this paper, we create a motor learning framework consisting of state-of-the-art components in motor skill learning and compare it to a manually designed program on the task of robot tetherball. We use dynamical motor primitives for representing the robot's trajectories and relative entropy policy search to train the motor framework and improve its behavior by trial and error. These algorithmic components allow for high-quality skill learning while the experimental setup enables an accurate evaluation of our framework as robot players can compete against each other. In the complex game of robot tetherball, we show that our learning approach outperforms and wins a match against a high quality hand-crafted system.

Authors

• Simone Parisi
• Hany Abdulsamad
• Alexandros Paraschos
• Christian G. Daniel
• Jan Peters 0001

Keywords

• Trajectory
• Robot kinematics
• Mathematical model
• Games
• Learning (artificial intelligence)
• Analytical models
• Dynamical
• Motor Skills
• Kullback-Leibler
• Challenges In Order
• Policy Search
• Parametrized
• Reward Function
• Markov Decision Process
• Policy Learning
• Pivot Point
• Function Of Force
• Low-level Control
• Joint Velocity
• Real Robot
• Policy Update
• Imitation Learning
• Number Of Basis Functions
• Gaussian Basis Function
• Real Learning
• Robot Learning
• Number Of Demonstrations
• Ball Position
• Gravity Compensation
• Ball Trajectory
• Ball Velocity
• Intercept Point
• Nonholonomic
• Lagrange Multiplier
• Sampling Efficiency
• Degrees Of Freedom