Learning manipulation actions from human demonstrations

Learning from demonstration is a popular approach for teaching robots as it allows service robots to acquire new skills without explicit programming. However, for manipulation actions mostly kinesthetic teaching is used as these actions require precise knowledge about the interactions between the robot and the object. In this paper, we present a novel approach that allows a robot to learn actions carried out by a teacher from observations. We achieve this by first transforming RGBD observations to consistent hand-object trajectories, which are then adapted to the robot's grasping capabilities. Experimental results show that the robot is able to learn complex tasks such as opening doors or drawers.

Authors

• Tim Welschehold
• Christian Dornhege
• Wolfram Burgard

Keywords

• Trajectory
• Grasping
• Motion segmentation
• Education
• Grippers
• Optimization
• Manipulation Actions
• Open Door
• Service Robots
• Inverse Reinforcement Learning
• Objective Function
• Mixture Model
• Object Detection
• Error Function
• Motion Model
• Manipulation Tasks
• Depth Camera
• Fisher Information
• Object Motion
• Scalar Product
• Actual Observations
• Human Hand
• Object Pose
• Object In Frame
• Graph Optimization
• Object Trajectory
• Hand Trajectory
• Robotic Gripper
• Motion Generation