Abstract In this paper, a Nonsingular Fast Terminal Sliding Mode Control (RNFTSMC) is proposed to regulate the position of a mobile lower limb exoskeleton powered by linear actuators with four degrees of freedom. The developed lower limb exoskeleton has the purpose to assist the sit-to-stand and stand-to-sit tasks for people with spinal cord injury (SCI) or elderly people. Currently most of the developed lower limb exoskeletons are dealing with the presence of uncertainties (unmodeled dynamics) and external disturbances (mass changes, opposite forces), caused by the high complexity of the human limbs dynamics. The RNFTSMC has the purpose of dealing with these uncertainties, ensuring a rapid convergence in finite time to the desired position, and reducing the chattering effect produced by the nature of the controller actuators. The human intention movements have been detected using an sEMG signal classifier performed by the Myo Armband device and processed to activate the preprogrammed routines in the RNFTSMC. The desired paths for the knee and hip exoskeleton joints were obtained by recording data from a volunteer who rises from a chair and sat on it, using a motion capture system. To evaluate the effectiveness of the proposed control scheme, real-time experiments under three different scenarios have been carried out.

中文翻译：

---

使用移动下肢外骨骼进行站立任务的鲁棒非奇异快速终端滑模控制

摘要 在本文中，提出了一种非奇异快速终端滑模控制（RNFTSMC）来调节由四自由度线性致动器驱动的移动下肢外骨骼的位置。开发的下肢外骨骼旨在协助脊髓损伤（SCI）或老年人的坐立和立坐任务。目前，大多数开发的下肢外骨骼都在处理由人类四肢动力学的高度复杂性引起的不确定性（未建模动力学）和外部干扰（质量变化、相反力）的存在。RNFTSMC 的目的是处理这些不确定性，确保在有限时间内快速收敛到所需位置，并减少控制器执行器的性质产生的颤振效应。使用由 Myo Armband 设备执行的 sEMG 信号分类器检测到人类意图运动，并对其进行处理以激活 RNFTSMC 中的预编程例程。膝关节和髋关节外骨骼关节所需的路径是通过使用运动捕捉系统记录一名从椅子上站起来并坐在椅子上的志愿者的数据获得的。为了评估所提出的控制方案的有效性，在三种不同场景下进行了实时实验。