论文部分内容阅读
针对具有非零初始误差的多关节机械手的轨迹跟踪控制问题,首先采用降阶变换的方式将原系统转化为低阶系统,再对低阶系统重新变换设计,该方法消除了Kawamura方法在学习过程中要求每次学习初态与期望初态一致或迭代初态固定的限制.然后提出一种新型快速迭代算法对变换后的机械手系统模型进行学习控制,充分利用了系统已存的有效信息,使系统的输出能尽快地收敛于期望轨迹.该算法相对于P型算法具有较快的跟踪收敛速度,相对于PD型算法计算过程简便易行,避免了微分运算给系统带来的不稳定影响.仿真结果表明了所提控制方案的有效性,提高了系统的实时性.
Aiming at the trajectory tracking control problem of multi-joint manipulator with non-zero initial error, the original system is transformed into a low order system by using a reduced order transformation and then the lower order system is re-transformed. This method eliminates the Kawamura method in the learning process Which requires that the initial state of each learning be consistent with the initial state or the initial state of iteration is fixed.Then a new fast iterative algorithm is proposed to study and control the transformed manipulator system model and make full use of the valid information stored in the system so that The output of the system can converge to the expected trajectory as soon as possible.The algorithm has faster tracking convergence speed than the P-type algorithm, and the calculation process is simple and easy compared with the PD-type algorithm, which avoids the instability brought by the differential operation to the system. The simulation results show the effectiveness of the proposed control scheme and improve the real-time performance of the system.