부산대학교 도서관

It is necessary to accurately and stably monitor the feedback signal during the bone milling process so that the surgical robot can better control the milling. In response to this problem, this article uses a vibration sensor fixed on the milling cutter for signal acquisition. During the drilling process, the milling cutter will produce different amplitudes of vibration under different conditions, the vibration signal collected by the acceleration sensor array is subjected to Fast Fourier Transform (FFT) and frequency domain features are extracted, and then the milling state is monitored through the BP neural network. Experiments show that the recognition rate of milling cortical bone and cancellous bone reaches 100%; Under the condition of cancellous bone, the recognition accuracy of the milling depth into three states of 0.5mm, 0.8mm and, 1.2mm are all above 100%. The comparison with a single sensor proves that the sensor array has better accuracy and stability in the robot milling process.