📄 中文摘要
深蹲是一种要求较高的下肢运动，需耗费大量肌肉力量和协调能力。通过智能化和个性化的辅助技术降低这一任务的身体需求具有重要意义，尤其是在涉及重复低强度组装活动的行业中。研究评估了一种模块化髋膝外骨骼的神经网络控制器在辅助深蹲任务中的有效性。该控制器通过强化学习在基于物理的人机交互仿真环境中进行训练，能够根据最近的关节角度和速度历史生成实时的髋关节和膝关节辅助扭矩。五名健康成年人参与了实验，结果表明该控制器能够显著降低深蹲时的肌肉负担。
基于强化学习的辅助技术通过模块化髋膝外骨骼减少深蹲负担
📄 English Summary
Reinforcement-Learning-Based Assistance Reduces Squat Effort with a Modular Hip--Knee Exoskeleton
Squatting is a highly demanding lower-limb movement that requires substantial muscular effort and coordination. Reducing the physical demands of this task through intelligent and personalized assistance has significant implications, particularly in industries involving repetitive low-level assembly activities. This study evaluated the effectiveness of a neural network controller for a modular Hip-Knee exoskeleton designed to assist squatting tasks. The controller was trained using reinforcement learning (RL) in a physics-based human-exoskeleton interaction simulation environment. It generated real-time hip and knee assistance torques based on recent joint-angle and velocity histories. Five healthy adults participated in the experiments, and the results indicated that the controller significantly reduced muscular effort during squatting.