基于自适应卡尔曼滤波的叉车质心侧偏角估计

Estimation of forklift sideslip angle using adaptive Kalman filters

质心侧偏角是叉车动力学研究中的重要状态参数,工程上常采用卡尔曼滤波器估算获取该参数值。文章以CPD15型电动叉车为研究对象,考虑到叉车质心以及前、后轮侧偏刚度会随货物的质量和形状产生较大的变化,建立了带有模型误差的线性二自由度叉车模型,基于该模型设计了自适应卡尔曼滤波器估计叉车质心侧偏角,并采用遗传算法在线优化滤波器参数,有效地解决了当叉车模型参数发生变化时卡尔曼滤波器估计精度降低的问题。Matlab仿真结果表明,自适应卡尔曼滤波器不仅能滤除叉车运动中的随机不确定性噪声,还能有效抑制未知的模型误差给估计带来的不利影响,增强了滤波器的估计精度。

Sideslip angle is a very important variable of forklift dynamics research,and Kalman filter is often used to estimate it.Taking CPD15-type forklift as the research object,and considering that the center of mass and tire force of the forklift often change with load,the linear two degree of freedom model with model error is established and the adaptive Kalman filter with online optimization based on genetic algorithm is designed to estimate forklift sideslip angle.The problem that the estimation accuracy of the classical Kalman filter is reduced with the varying parameters of the forklift is effectively solved.Then the Matlab simulation test is done,and the simulation result shows that the adaptive Kalman filter can not only filter the uncertain random noise of the system,but also effectively restrict the adverse effects of the uncertain model parameter error when forklift is working.Meanwhile,the estimation accuracy of the filter is effectively enhanced.