A kernel-based approach to direct action perception

The direct perception of actions allows a robot to predict the afforded actions of observed objects. In this paper, we present a non-parametric approach to representing the affordance-bearing subparts of objects. This representation forms the basis of a kernel function for computing the similarity between different subparts. Using this kernel function, together with motor primitive actions, the robot can learn the required mappings to perform direct action perception. The proposed approach was successfully implemented on a real robot, which could then quickly learn to generalize grasping and pouring actions to novel objects.

Authors

• Oliver Kroemer
• Emre Ugur
• Erhan Öztop
• Jan Peters 0001

Keywords

• Robots
• Kernel
• Grasping
• Trajectory
• Humans
• Shape
• Visualization
• Kernel Function
• Real Robot
• Physical Interaction
• Scope Of This Paper
• Visual Features
• Point Cloud
• Weight Vector
• Weight Function
• Optical Flow
• Goal State
• Similar Objects
• Present Task
• Plastic Cups
• Surface Distribution
• Small Water
• Surface Geometry
• Length Scale Parameter
• Kernel Values
• Object Point Cloud
• Time-of-flight Sensors
• Object Affordances