Abstract
Achieving a comprehensive protection system associated with unpredictability of human behavior, complexity of the cutting task and variety of the environmental working condition, is a task which presents a considerable challenge. To address such difficulties, this paper presents an effective detection algorithm based on the measurement of impedance between the bar of chainsaw and the rear handle, extended by the human body. A multitone signal has been exploited to evaluate the system's response and validate the equivalent electrical model experimentally. To evaluate the transfer function, discrete Fourier Transform is performed, and the Fourier coefficients are evaluated by multiplying the acquired signals by two vectors of sine and cosine at four frequencies. As results show, the magnitude of the transfer function carries information about the approaching distance while phase provides very useful information to distinguish between the proximity to body instead of tree. Finally, the developed algorithm demonstrates a successful human body detection rate of 100% with 4% of false-positive detections.
Funding Information
  • Redcap S.r.l
  • Made in Italy–Circular and Sustainable (MICS) Extended Partnership
  • Next-Generation EU (CUP MICS D43C22003120001)

This publication has 30 references indexed in Scilit: