Noise Processing Method of MEMS Tilt Sensor using Improved Kalman Filter based on Quantum Particle Swarm Optimization

This paper introduces a novel approach combining quantum particle swarm optimization (QPSO) and Kalman filter to enhance the anti-noise performance of micro-electro-mechanical system (MEMS) tilt sensors, which are susceptible to external environmental noise interference, affecting their measurement accuracy and reliability. The parameters of Kalman filter are adaptively optimized by QPSO, which addresses the issues of local optima and premature convergence during the optimization and correction process of prediction system. Then Kalman filter is carried out with the optimal parameters to achieve the effect of denoising. To validate the denoising performance of proposed algorithm, an experimental scene was set up with a six-dimensional space vibration test bench. According to the experimental findings, the proposed method exhibits a superior noise reduction effect, as evidenced by its smaller mean absolute error (MAE) and mean square error (MSE) compared to alternative techniques such as variational mode decomposition (VMD) combined with wavelet transform (WT), back propagation (BP) neural network optimized Kalman filter and particle swarm optimization (PSO) improved Kalman filter. The contribution lies in the innovative integration of QPSO with Kalman filtering, addressing a critical need in engineering by fortifying the anti-noise capabilities of MEMS tilt sensors, thereby bolstering their measurement accuracy and reliability in challenging environmental conditions.