Nickel-titanium (Ni-Ti) alloys are characterized by unique mechanical properties including superelasticity, high ductility, and severe strainhardening,
that make them extremely difficult to cut. In this paper, in order to realize a reliable and robust classification of process conditions, a
multiple sensor monitoring system is employed to acquire cutting force and vibration acceleration sensor signals during experimental turning
tests on Ni-Ti alloys. The acquired sensorial data were subjected to an advanced sensor signal processing procedure based on signal spectral
estimation allowing for feature extraction from the signal frequency content. The extracted features were utilised to build both single signal
component feature vectors as well as sensor fusion feature vectors to be fed as input to adaptive neuro-fuzzy systems for pattern recognition,
with the aim to investigate the correlation between the input pattern feature vectors and the output process quality