双E型弹性体六维力传感器薄矩形板AWGN-Kalman滤波

AWGN-Kalman Filter for Thin Rectangular Board of Dual-E Elastomer Six-axis Force Sensor

针对六维力传感器采集信号中存在复杂噪声干扰的问题,为有效地对系统降噪和提高传感器的分辨率,本文以六维力传感器薄矩形板为研究对象,引入应变片热噪声与信号处理电路的散粒噪声信号,分析其窄带高斯统计特性,融入系统状态方程,生成状态加性高斯白噪声﹒借鉴Box-Muller变换法,验证了状态噪声的高斯统计特性,进而构建了线性离散时间系统状态模型,并详细推导了AWGN-Kalman滤波公式﹒通过EKF和UKF算法滤波实例,验证了该模型降噪的有效性,且选择合适的滤波方法,能更有效地对六维力传感器的薄矩形板进行滤波.

To solve the problem of complex noise interference in the acquisition signal of the six-axis force sensor and to effectively reduce the noise of the system and improve the resolution of the sensor,this paper takes the thin rectangular plate of the six-axis force sensor as the research object.The thermal noise of strain gauge and shot noise of signal processing circuit are introduced.After analyzing the narrow-band Gaussian statistical characteristics,the system state equation is integrated to generate state additive Gaussian white noise.Based on the Box-Muller transform method,the Gaussian statistical characteristics of state noise are verified,then the state model of linear discrete-time system is constructed,and the AWGN-Kalman filtering formula is deduced in detail.The example shows that the EKF and UKF filtering algorithm verify the effectiveness of noise reduction of the model,and the choice of the appropriate filtering method can more effectively filter the thin rectangular plate of six-axis force sensor.