SkiP: When to Skip and When to Refine for Efficient Robot Manipulation

TL;DR

SkiP is a novel robot manipulation policy that intelligently skips redundant steps and refines key actions, significantly reducing execution steps while maintaining or improving success rates.

Contribution

The paper introduces SkiP, a unified policy that dynamically skips non-essential steps and refines critical actions without hierarchical planning, along with MSK for automatic segmentation.

Findings

Reduces executed steps by 15-40% across tasks.

Matches or improves success rates compared to baseline policies.

Works with various policy architectures.

Abstract

Previous imitation learning policies predict future actions at every control step, whether in smooth motion phases or precise, contact-rich operation phases. This uniform treatment is wasteful: most steps in a manipulation trajectory traverse free space and carry little task-relevant information, while a small fraction of \emph{key} steps around contacts, grasps, and alignment demand dense, high-resolution prediction. We propose a novel \emph{action relabeling} mechanism: at each timestep in a skip segment, we replace the behavior cloning target with the action at the entrance of the next key segment, enabling the policy to leap over redundant steps in a single decision. The resulting \textbf{Skip Policy (SkiP)} dynamically leaps over skip segments and intensively refines actions in key segments, within a single unified network requiring no learned skip planner or hierarchical structure. To automatically partition demonstrations into key and skip segments without manual annotation, we introduce \emph{Motion Spectrum Keying} (MSK), a fast, task-agnostic procedure that detects local motion complexity from action signals. Extensive experiments across 72 simulated manipulation tasks and three real-robot tasks show that SkiP reduces executed steps by $15$--$40\%$ while matching or improving success rates across various policy backbones. Project page: \texttt{https://pgq18.github.io/SkiP-page/}.