This paper describes the development and evaluation of a newly developed end-effector type rehabilitation robot named DDgo Pro. In order to induce normal walking pattern, a five-link mechanism was designed to follow normal walking trajectory. In order to implement the rehabilitation training feature, the core technologies of the drive unit consisting of freewheel and BLDC motor, the construction of the robot control system, and the algorithms of the three training modes in DDgo Pro were addressed in detail. Among the three training modes, Passive Mode makes the robot to fully guide normal walking pattern, and to help patients gain muscle strength in their lower body. In Active Assisted Mode, patients who have learned Passive Mode are able to perform active rehabilitation training for a long time while receiving muscle assistance from the robot in proportion to their walking intentions. Active mode was designed for patients to perform rehabilitation training for themselves with minimal help from the robot. In order to test muscle assisted performance of the newly developed rehabilitation robot, muscle activity in each training mode was measured using Electromyography for healthy people. In addition, gait dynamics was analyzed to check whether DDgo Pro help stroke patients correct their walking patterns.

本文介绍了一种新开发的名为DDgo Pro的末端执行器式康复机器人的开发和评估。为了诱发正常的步行模式，设计了一种五连杆机构来遵循正常的步行轨迹。为了实现康复训练功能，详细介绍了由飞轮和BLDC电机组成的驱动单元的核心技术，机器人控制系统的结构以及DDgo Pro中三种训练模式的算法。在这三种训练模式中，被动模式使机器人能够完全引导正常的步行模式，并帮助患者获得下半身的肌肉力量。在主动辅助模式下，已学习被动模式的患者能够长时间进行主动康复训练，同时根据机器人的步行意图接受机器人的肌肉帮助。主动模式旨在让患者在机器人的帮助下进行自我康复训练。为了测试新开发的康复机器人的肌肉辅助性能，使用肌电图仪对健康人测量了每种训练模式下的肌肉活动。此外，还对步态动力学进行了分析，以检查DDgo Pro是否有助于中风患者纠正其行走方式。