【摘要】 在智能车实际运动特性优化控制的研究中,由于传统非线性算法难以适应无人驾驶智能车实际运动特点的问题,提出一种基于交互式多模型(IMM)带衰减因子的自适应无迹卡尔曼滤波(UKF)方法。上述方法在UKF基础上增加了衰减因子、自适应滤波理论、IMM结构。克服了历史数据对滤波的影响、系统线性化误差、以及参数设置难以适应模型不确定性等技术难点。仿真结果表明:改进方法比传统滤波方法的估计精度高,稳定性好,收敛速度快而且鲁棒性更强,满足了智能车导航系统的实际用需求。采用改进方法的智能车导航控制系统提高了无人驾驶智能车整体的安全性和可靠性。更多还原

【Abstract】 In view of the traditional nonlinear algorithm is difficult to adapt to actual motion BJUT-IV unmanned intelligent vehicle in target tracking,a new adaptive unscented Kalman filter based on the interacting multiple model with fading factor algorithm( IMM-AFUKF) is proposed. The IMM-AFUKF algorithm adds the fading factor,adaptive filter theory and IMM structure to solve the problem of UKF. Simulation results show that,compared with the standard algorithm,the proposed algorithm provides better accuracy,stability and convergent rate. By using IMM-AFUKF algorithm,the navigation system has higher accuracy and reliability than traditional algorithm.更多还原