Stage-Aware and Roughness-Constrained Diffusion Policy for Multi-Stage Robotic Polishing

2026-06-24

AI Summary

This work addresses the challenges of long-horizon dependencies, uncertain phase transitions, and strong coupling among process parameters in robotic polishing by proposing a phase-aware, roughness-constrained diffusion policy. The method leverages multimodal historical observations to infer polishing phases in an unsupervised manner and guides action generation within a conditional diffusion framework. A roughness-guided constrained sampling mechanism is introduced to jointly modulate feed rate and contact force in alignment with a prescribed spindle speed. By integrating phase awareness and physical constraints into the diffusion policy for the first time, the approach enables coherent action generation and coordinated parameter control without requiring explicit phase labels. Evaluations on spacecraft cabin coating and internal cavity polishing tasks demonstrate significant improvements in phase transition stability, parameter consistency, and surface finish quality.

Abstract

Polishing is a critical finishing process in high-end manufacturing fields such as aerospace, where surface quality directly affects the service performance and reliability of components. Robotic imitation learning provides a flexible solution for such tasks, but current methods remain limited in industrial polishing because of long-horizon dependencies, uncertain stage transitions, and the difficulty of modeling and regulating coupled process parameters. To address these issues, this paper proposes a Stage-Aware and Roughness-Constrained Diffusion Policy (SRDP) for robotic polishing. SRDP infers the process-stage posterior from multimodal observation histories and uses it to condition the shared reverse denoising process, enabling stage-consistent action generation without external stage labels during execution. Furthermore, a roughness-oriented process-constrained diffusion sampling method is incorporated to generate constrained feed speed and normal contact force under stage-wise preset spindle speeds, thereby improving process consistency and physical feasibility. Systematic experiments are conducted on two representative scenarios, namely spacecraft cabin coating-surface polishing and inner-cavity structural surface finishing. Comparisons with advanced baselines, ablation studies, and real-robot validations comprehensively evaluate the proposed method. The results show that SRD improves stage-transition stability, process-parameter consistency, and final surface quality across different polishing scenarios.

Problem

• robotic polishing
• long-horizon dependencies
• stage transitions
• process parameters
• surface quality

Innovation

• Stage-Aware Diffusion Policy
• Roughness-Constrained Sampling
• Robotic Polishing
• Process Consistency
• Imitation Learning