Analysis of Force Signals for the Estimation of Surface Roughness during Robot-Assisted Polishing.

Beatriz De Agustina,Roberto Teti,Eva Rubio,Marta Marín

doi:10.3390/ma11081438

Beatriz De Agustina, Roberto Teti + Show 2 more

Open Access

https://doi.org/10.3390/ma11081438

Copy DOI

Abstract

In this study feature extraction of force signals detected during robot-assisted polishing processes was carried out to estimate the surface roughness during the process. The purpose was to collect significant features from the signal that allow the determination of the end point of the polishing process based on surface roughness. For this objective, dry polishing turning tests were performed on a Robot-Assisted Polishing (RAP) machine (STRECON NanoRAP 200) during three polishing sessions, using the same polishing conditions. Along the tests, force signals were acquired and offline surface roughness measurements were taken at the end of each polishing session. As a main conclusion, it can be affirmed, regarding the force signal, that features extracted from both time and frequency domains are valuable data for the estimation of surface roughness.

Highlights

These days, finishing processes such as polishing are not yet fully automated in manufacturing industries
A considerable percent of the total time required for manufacturing products such as dies, molds, machine tools, and optical components is spent on the finishing operations, which account for approximately 30% to 50% of the manufacturing time [2,3]
Due to the fact that polishing is not performed by position control but by pressure control, efforts towards automated polishing have mainly focused on the identification and the execution of the motions involved in manual polishing techniques, performing under a desired contact force exerted on the tool [4]. This is an important challenge in die polishing, especially for the manufacturing of components with complex shapes and machine tools that contain freeform surfaces and functional relevant edges

Summary

Introduction

These days, finishing processes such as polishing are not yet fully automated in manufacturing industries. Polishing has largely been a manual operation that is very labor-intensive, highly skill-dependent, inefficient due to long processing times, high-cost, error-prone, and hazardous due to abrasive dust [1] For this reason, a considerable percent of the total time required for manufacturing products such as dies, molds, machine tools, and optical components is spent on the finishing operations, which account for approximately 30% to 50% of the manufacturing time [2,3]. Due to the fact that polishing is not performed by position control but by pressure control, efforts towards automated polishing have mainly focused on the identification and the execution of the motions involved in manual polishing techniques, performing under a desired contact force exerted on the tool [4] This is an important challenge in die polishing, especially for the manufacturing of components with complex shapes and machine tools that contain freeform surfaces and functional relevant edges. Force control methods have been designed and can be broadly classified as active and passive force controls

Methods

Results

Conclusion