小波分析在机器人腕力传感器信号去噪中的应用

机器人腕力传感器在工业现场测量力或力矩时，不可避免的受到随机噪声的干扰，从而影响了测量精度的提高。为了克服传统去噪方法的局限性，本文将多重小波变换应用到机器人腕力传感器信号去噪中，采用浮动阈值法消除噪声，并将传统的低通滤波和FFT／IF丌方法与本文介绍的方法进行比较。结果表明，多重小波浮动阈值去噪方法在机器人腕力传感器信号去噪方面优于传统的方法。

Application of Multi-wavelet in Data Pretreatment of Robot Wrist Force Sensor

During robot wrist force sensor measures force/Torque, the signals are inevitably influenced by stochastic noises, which have poor effects on its measure precision. In order to filter stochastic noises, compared with low pass filter and FFT/IFFT, the theory of multi-wavelet is analyzed and applied to process the data measured by robot wrist force sensor, and the method of soft threshold is adopted during the data process. The experimental results show that the proposed method is effective for eliminating the effect of stochastic noises, but the performance of the new method excels that of the conventional method.