Composable Interaction Primitives: A Structured Policy Class for Efficiently Learning Sustained-Contact Manipulation Skills

We propose a new policy class, Composable Interaction Primitives (CIPs), specialized for learning sustained-contact manipulation skills like opening a drawer, pulling a lever, turning a wheel, or shifting gears. CIPs have two primary design goals: to minimize what must be learned by exploiting structure present in the world and the robot, and to support sequential composition by construction, so that learned skills can be used by a task-level planner. Using an ablation experiment in four simulated manipulation tasks, we show that the structure included in CIPs substantially improves the efficiency of motor skill learning. We then show that CIPs can be used for plan execution in a zero-shot fashion by sequencing learned skills. We validate our approach on real robot hardware by learning and sequencing two manipulation skills.

Authors

• Ben Abbatematteo
• Eric Rosen
• Skye Thompson
• Tuluhan Akbulut
• Sreehari Rammohan
• George Konidaris 0001

Keywords

• Sequential analysis
• Vocabulary
• Visualization
• Wheels
• Turning
• Motors
• Hardware
• Efficient Learning
• Interaction Primitives
• Motor Skills
• Learning Skills
• Manipulation Tasks
• Ablation Experiments
• Deep Network
• Point Cloud
• Path Planning
• Task Order
• Executive Skills
• Joint State
• Inverse Kinematics
• Safe Learning
• Joint Limits
• Joint Configuration
• Safety Constraints
• Touch Sensor
• Sustained Contact
• Upper Confidence Bound
• Bandit Problem
• End-effector Pose
• Violation Rate
• Contact Gap
• Torque Limits
• Feedback Control
• Sensor Inputs
• Shape Of Trajectory
• Motor Control