Acoustic Feedback for Closed-Loop Force Control in Robotic Grinding
Zongyuan Zhang, Christopher Lehnert, Will Browne, Jonathan Roberts
AI summary
Problem
Robotic grinding systems rely heavily on costly and hard-to-adapt force sensors for closed-loop control, while ignoring readily available acoustic signals that correlate with grinding dynamics.
Approach
The authors propose a data-driven system that uses a single contact microphone and a 2D convolutional neural network to estimate normal grinding force from audio in real time, feeding it directly into a hybrid force-position controller.
Key results
- 4-fold increase in Material Removal Rate consistency across varying disc conditions
- Real-time normal force estimation from raw audio using a tailored 2D CNN
- Robust acoustic force estimation under ambient noise
- Approximately 200-fold cost reduction compared to conventional force sensors
Why it matters
Enables affordable, easily deployable robotic grinding for manufacturing and fabrication by eliminating the need for expensive force sensing hardware.
Abstract
Acoustic feedback is a critical indicator for assess- ing the contact condition between the tool and the workpiece when humans perform grinding tasks with rotary tools. In con- trast, robotic grinding systems typically rely on force sensing, with acoustic information largely ignored. This reliance on force sensors is costly and difficult to adapt to different grinding tools, whereas audio sensors (microphones) are low-cost and can be mounted on any medium that conducts grinding sound. This paper introduces a low-cost Acoustic Feedback Robotic Grinding System (AFRG) that captures audio signals with a contact microphone, estimates grinding force from the audio in real time, and enables closed-loop force control of the grinding process. Compared with conventional force-sensing approaches, AFRG achieves a 4-fold improvement in consistency across different grinding disc conditions. AFRG relies solely on a low- cost microphone, which is approximately 200-fold cheaper than conventional force sensors, as the sensing modality, providing an easily deployable, cost-effective robotic grinding solution.