Push-Wiper: Toward General-Purpose Robotic Cleaning across Varied Stains and Surfaces with Segmented Pushing Trajectories

2026-08-01

AI Summary

This work addresses the challenge of removing highly viscous stains, whose complex rheology often causes conventional cleaning methods to either spread the contaminant or damage the surface. The study introduces a novel formulation of stain removal as an aggregation task and proposes a segmented pushing trajectory combined with a self-cleaning mechanism. Leveraging a Diffusion Policy to generate adaptive pushing actions, the approach integrates an Arbitrary Surface Pose Interpolator (ASPI) and a hybrid force-position controller to generalize across diverse spatial stain distributions. Without additional training, the method achieves zero-shot transfer to multiple unseen scenarios—including solid residues, liquid spills, stains of unknown viscosity, and curved surfaces—outperforming baseline approaches by up to 130% in cleaning score (CS).

Abstract

Viscous stains, characterized by high viscosity and complex rheological properties, remain a major challenge for robotic surface cleaning. Conventional wiping often spreads the stain, while scrubbing provides stronger friction but risks damaging the surface. In this paper, we propose Push-Wiper, a framework that reformulates viscous stain cleaning as an aggregation problem. Push-Wiper employs a sponge to progressively gather stains through segmented pushing trajectories, followed by a post-processing phase that detaches the aggregated material and enables sponge self-cleaning. We adopt a stepwise strategy for stain gathering and leverage Diffusion Policy to generate adaptive pushing action sequences. These sequences are executed through our Arbitrary Surface Pose Interpolator (ASPI) and a hybrid force-position controller, allowing the method to generalize to stains with diverse spatial distributions. Push-Wiper achieves a cleaning score (CS), defined as the percentage of stain area removed, up to 130% higher than baseline methods. Without additional training, Push-Wiper also transfers in a zero-shot manner to solid residues, liquid spills, unseen viscous stains, and curved surfaces with varying geometries. Our experiments demonstrate the cleaning effectiveness of Push-Wiper and its strong generalization ability. The project website is available at https://push-wiper.github.io/.

Problem

• viscous stains
• robotic cleaning
• surface cleaning
• generalization
• cleaning automation

Innovation

• Push-Wiper
• segmented pushing trajectories
• Diffusion Policy
• zero-shot generalization
• viscous stain cleaning