基于ESO的OMR轨迹跟踪预测控制

Trajectory Tracking Predictive Control for Omnidirectional Mobile Robots Based on Extended State Observer

针对麦克纳姆轮全向移动机器人实际系统模型中广泛存在的未建模动态、非线性摩擦和外界未知干扰等显著影响轨迹跟踪控制性能的不确定性,提出基于扩张状态观测器的广义预测控制。将所有干扰看做一个扩张状态量,通过对扰动量进行在线估计并实时补偿,提高了轨迹跟踪控制器的抗干扰能力。与基于卡尔曼滤波的广义预测控制器进行对比,仿真结果表明,所提出的方法可以有效抑制参数摄动和未知干扰的影响,提高麦克纳姆轮全向移动机器人轨迹跟踪控制器抗干扰性能。

Aiming at the uncertainties of the trajectory tracking control performance,such as unmodeled dynamics,nonlinear friction,and external unknown disturbance,which widely exist in the actual system model of the mecanum wheel omnidirectional mobile robot,a generalized predictive control based on extended state observer is proposed.The disturbance was regarded as an extended state variable,and the disturbance momentum was estimated online and compensated in real-time.The anti-interference ability of the trajectory tracking controller was improved.Compared with the generalized predictive controller based on the Kalman filter,the simulation results show that the proposed method can effectively suppress the influence of parameter perturbation and unknown disturbance,and improve the anti-interference performance of the trajectory tracking controller for the mecanum wheel omnidirectional mobile robot.