基于带约束DUKF和AMPC的喷雾机路径跟踪控制  

Path Tracking Control of Sprayer Based on Constrained DUKF and AMPC

作　　者：郑靖科 戴海峰[1] 

机构地区：[1]同济大学新能源汽车工程中心,上海201804

出　　处：《机电一体化》2022年第6期3-12,21,共11页Mechatronics

摘　　要：为提高喷雾机在自身参数变化和复杂工作环境下的自动循迹精度,提出了一种基于带约束DUKF和AMPC的路径跟踪控制方法。带约束的DUKF包括两个无迹卡尔曼滤波器,分别用来估计喷雾机的状态和参数。AMPC路径跟踪控制器根据实时获取到的估计值自适应修正喷雾机动力学模型,以保证在参数变化和受到外部干扰时,仍然能够实现对参考路径的精确跟踪。仿真和试验结果表明,在典型的U形全覆盖参考路径下,参数估计误差不超过7.5%,喷雾机自动循迹的最大误差不超过0.05 m,验证了提出方法的可行性和优越性。In order to improve the path tracking accuracy of spraying machine under parameters variation and complex working condition, a path tracking control method based on constrained dual unscented Kalman filters(DUKF) and adaptive model predictive control(AMPC) was proposed. The constrained DUKF was composed of two unscented Kalman filters to estimate the state and parameters of the spraying machine, respectively. Its dynamical model was modified adaptively by the MPC path tracking controller according to the estimated parameters obtained in real time to ensure high precision of path tracking under parameters variation and external disturbances. The results of simulation and field test showed that the parameter estimation error was less than 7.5%, and the maximum error of path tracking was less than 0.05 m, which verified the feasibility and superiority of the proposed method.

关 键 词：无迹卡尔曼滤波 模型预测控制 路径跟踪 参数估计 

分 类 号：S491[农业科学—植物保护] TP273[自动化与计算机技术—检测技术与自动化装置]