Learning robot tactile sensing for object manipulation

Tactile sensing is a fundamental component of object manipulation and tool handling skills. With robots entering unstructured environments, tactile feedback also becomes an important ability for robot manipulation. In this work, we explore how a robot can learn to use tactile sensing in object manipulation tasks. We first address the problem of in-hand object localization and adapt three pose estimation algorithms from computer vision. Second, we employ dynamic motor primitives to learn robot movements from human demonstrations and record desired tactile signal trajectories. Then, we add tactile feedback to the control loop and apply relative entropy policy search to learn the parameters of the tactile coupling. Additionally, we show how the learning of tactile feedback can be performed more efficiently by reducing the dimensionality of the tactile information through spectral clustering and principal component analysis. Our approach is implemented on a real robot, which learns to perform a scraping task with a spatula in an altered environment.

Authors

• Yevgen Chebotar
• Oliver Kroemer
• Jan Peters 0001

Keywords

• Tactile sensors
• Trajectory
• Vectors
• Kernel
• Estimation
• Tactile Sensor
• Manipulation Tasks
• Pose Estimation
• Altered Environment
• Spectral Clustering
• Tactile Information
• Robot Movement
• Relative Search
• Tactile Signals
• Policy Search
• Gaussian Kernel
• Point Cloud
• Nonlinear Dynamics
• Task Execution
• Leave-one-out Cross-validation
• Number Of Weights
• Elastography
• Similar Appearance
• Object Parts
• Iterative Closest Point
• Voting Scheme
• Policy Update
• Object Pose
• Robotic Hand
• Canonical System
• Imitation Learning
• Feedback Parameter
• Function Of Force
• Observable Quantities