基于Backstepping的机器人鲁棒控制律设计

Backstepping - based Robust Control Scheme for Robot

考虑了仅有关节位置测量的机器人控制问题，首先设计滑模观测器估计速度信号，然后基于Ｂａｃｋｓｔｅｐｐｉｎｇ 的非线性控制系统综合方法，提出了一种机器人的鲁棒控制方案，可保证跟踪误差的一致最终有界．最后对所提出的鲁棒控制方案与常用的ＰＩＤ（比例积分微分）＋前馈补偿控制方案作了实验比较，结果表明，所提出的鲁棒控制方案具有更高的跟踪精度．

Considering the fact that there is only joint position measurement in most industrial robots, we have designed a sliding mode observer to estimate the velocities, then based on backsteping method, a robust control scheme for robot is proposed, which can guarantee that the tracking errors are uniformly and ultimately bounded. Finally we compared the robust control law with PID + forward feed compensation control law on a robot in experiments, which shows that the former control law is of smaller tracking error.