Exploration-assisted Bottleneck Transition Toward Robust and Data-efficient Deformable Object Manipulation

2026-03-14

AI Summary

This work addresses the performance degradation of imitation learning in deformable object manipulation caused by out-of-distribution (OOD) states, such as severe self-occlusion. The authors propose ExBot, a framework that introduces standardized “bottleneck states” as canonical starting points for task execution, thereby reframing OOD control as a state-transfer problem from arbitrary initial configurations to these bottleneck states. By integrating state-identifiability-based partitioning of the OOD space, dual action-primitive policies, and an active exploration mechanism, ExBot achieves robust manipulation without requiring high-fidelity perception. Experimental results demonstrate that ExBot efficiently handles diverse OOD initial conditions in real-world rope and cloth manipulation tasks, significantly improving both data efficiency and task success rates.

Technology Category

• Intelligent Robots: Manipulation
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments

Abstract

Imitation learning has demonstrated impressive results in robotic manipulation but fails under out-of-distribution (OOD) states. This limitation is particularly critical in Deformable Object Manipulation (DOM), where the near-infinite possible configurations render comprehensive data collection infeasible. Although several methods address OOD states, they typically require exhaustive data or highly precise perception. Such requirements are often impractical for DOM owing to its inherent complexities, including self-occlusion. To address the OOD problem in DOM, we propose a novel framework, Exploration-assisted Bottleneck Transition for Deformable Object Manipulation (ExBot), which addresses the OOD challenge through two key advantages. First, we introduce bottleneck states, standardized configurations that serve as starting points for task execution. This enables the reconceptualization of OOD challenges as the problem of transitioning diverse initial states to these bottleneck states, significantly reducing demonstration requirements. Second, to account for imperfect perception, we partition the OOD state space based on recognizability and employ dual action primitives. This approach enables ExBot to manipulate even unrecognizable states without requiring accurate perception. By concentrating demonstrations around bottleneck states and leveraging exploration to alter perceptual conditions, ExBot achieves both data efficiency and robustness to severe OOD scenarios. Real-world experiments on rope and cloth manipulation demonstrate successful task completion from diverse OOD states, including severe self-occlusions.

Problem

• Deformable Object Manipulation
• Out-of-Distribution
• Imitation Learning
• Self-occlusion
• Data Efficiency

Innovation

• bottleneck states
• exploration-assisted manipulation
• deformable object manipulation
• out-of-distribution robustness
• dual action primitives