Dynamic modeling and chatter detection in high-speed robotic milling using multi-sensor fusion

Abstract

In this work, a dynamic model of the robotic system is established, and its stability is analyzed using the zero-order approximation method. The characteristic equation is solved to predict stability, and a stability lobe diagram is generated to guide the selection of experimental parameters. To achieve a comprehensive analysis of chatter, multi-sensor fusion integrates AE signals and force signals to leverage their respective strengths. Root mean square values are utilized for effective stability monitoring, while fast Fourier transform (FFT) and short-time Fourier transform (STFT) provide detailed analyses of machining results to accurately identify chatter. Additionally, the machined surface morphology is examined and compared. By integrating the stability map with experimental results, the findings demonstrate that the proposed model and stability prediction method enable a more precise prediction of process stability. Furthermore, the combined analysis of acoustic emission signals, including time-domain and frequency-domain evaluations alongside force signal analysis, offers an effective approach for detecting chatter in robotic milling. This methodology provides valuable insights for optimizing machining parameters, thereby enhancing machining accuracy and efficiency.