Chris Paxton

Latest

• Learning Perceptual Concepts by Bootstrapping from Human Queries
• Correcting Robot Plans with Natural Language Feedback
• IFOR: Iterative Flow Minimization for Robotic Object Rearrangement
• HandoverSim: A Simulation Framework and Benchmark for Human-to-Robot Object Handovers
• Model Predictive Control for Fluid Human-to-Robot Handovers
• StructFormer: Learning Spatial Structure for Language-Guided Semantic Rearrangement of Novel Objects
• A Persistent Spatial Semantic Representation for High-level Natural Language Instruction Execution
• Language Grounding with 3D Objects
• Predicting Stable Configurations for Semantic Placement of Novel Objects
• SORNet: Spatial Object-Centric Representations for Sequential Manipulation
• NeRP: Neural Rearrangement Planning for Unknown Objects
• Alternative Paths Planner (APP) for Provably Fixed-time Manipulation Planning in Semi-structured Environments
• Reactive Human-to-Robot Handovers of Arbitrary Objects
• \"Good Robot\": Efficient Reinforcement Learning for Multi-Step Visual Tasks
• Collaborative Interaction Models for Optimized Human-Robot Teamwork
• Human Grasp Classification for Reactive Human-to-Robot Handovers
• “Zero-shot” Task Execution through Learning of Pixel-Based Action Preconditions
• 6-DOF Grasping for Target-driven Object Manipulation in Clutter
• Motion Reasoning for Goal-Based Imitation Learning
• Online Replanning in Belief Space for Partially Observable Task and Motion Problems
• Representing Robot Task Plans as Robust Logical-Dynamical Systems
• The CoSTAR Block Stacking Dataset: Learning with Workspace Constraints
• Conditional Driving from Natural Language Instructions