This article presents a control scheme for flexible joint robots which has uncertain parameters based on adaptive fuzzy compensation. Considering the unknown parameters, the proposed state feedback control approach utilizes measured variables to establish a cascade structure that is based on simplified dynamics. After reducing the number of fuzzy rules, the adaptive fuzzy logic system is added as compensation to decrease the approximated errors, and the robust terms are also used to enhance the robustness of closed-loop system. Then, the global asymptotic stability could be confirmed through Lyapunov stability principle and Barbalat's lemma. Compared with the other two controllers, the proposed control method has not only higher position accuracy and better dynamic performance but also robustness to the approximation of motor inertia, friction torque and link torque. Some simulation experiments are conducted to show the validity of the proposed scheme.

中文翻译：

---

基于不确定参数自适应模糊补偿的柔性关节机器人跟踪控制

本文提出了一种基于自适应模糊补偿的具有不确定参数的柔性关节机器人控制方案。考虑到未知参数，所提出的状态反馈控制方法利用测量变量来建立基于简化动力学的级联结构。在减少模糊规则的数量后，加入自适应模糊逻辑系统作为补偿，减少近似误差，同时利用鲁棒项增强闭环系统的鲁棒性。然后，通过Lyapunov稳定性原理和Barbalat引理可以证实全局渐近稳定性。与其他两种控制器相比，所提出的控制方法不仅具有更高的位置精度和更好的动态性能，而且对电机惯量的逼近具有鲁棒性，摩擦力矩和连杆力矩。进行了一些仿真实验以证明所提出方案的有效性。