This paper presents a model-free finite-time terminal sliding mode control scheme of uncertain robot systems. Time delay estimation (TDE) technique is employed to estimate mathematical model of system. However, TDE cannot achieve satisfactory performance when the system suffers from impactive disturbances. For this reason, a novel adaptive sliding mode observer is designed to compensate estimation errors. Moreover, the proposed observer is finite-time stable and the design process does not depend on the upper bounds of uncertainties and its derivatives, and it can be also extended to other nonlinear systems with external disturbance. A nonsingular fast terminal sliding surface is designed to accelerate convergence rate and boundary layer technique is adopted to attenuate the chattering phenomenon. Finally, a finite-time stable controller is constructed to stabilize the closed-loop system. The proposed controller is model-free and can be easily implemented in practice. The stability of both observer and controller is strictly proven in the Lyapunov framework. Simulations and experimental studies using Rethink Sawyer Robot are carried out to verify the effectiveness of the proposed scheme.

中文翻译：

---

具有不确定机器人系统自适应滑模观测器的无模型有限时间终端滑模控制

本文提出了一种不确定机器人系统的无模型有限时间终端滑模控制方案。时延估计（TDE）技术用于估计系统的数学模型。然而，当系统遭受冲击扰动时，TDE 无法获得令人满意的性能。为此，设计了一种新颖的自适应滑模观测器来补偿估计误差。此外，所提出的观测器是有限时间稳定的，设计过程不依赖于不确定性的上界及其导数，还可以扩展到其他具有外部干扰的非线性系统。设计了一个非奇异的快速终端滑动面以加快收敛速度​​，并采用边界层技术来衰减颤振现象。最后，构造了一个有限时间稳定控制器来稳定闭环系统。所提出的控制器是无模型的，并且可以在实践中轻松实现。观察者和控制器的稳定性在 Lyapunov 框架中得到严格证明。使用 Rethink Sawyer Robot 进行了模拟和实验研究，以验证所提出方案的有效性。