Research On The Prototype Of A Body Weight Support Gait Training Robot

There is growing clinical evidence that body weight support (BWS) gait training appears to be an effective intervention for enhancing locomotor recovery of lower limbs. However, there are still some problems, such as heavy workload to therapists,low efficiency, which prevent it from clinical application. Robot-aided gait training with partial body weight support may be a promising approach. This project, funded by the National Natural Science Foundation of China (No. 50575122) and the State Science and Technology Support Plan (2006BAI22B03), based on the former research of our group, aims to develop a prototype of body weight support gait training robot and then implement primary performance testing and preliminary application testing. The main contributions of this dissertation are shown as follows:1. Mechanism design of prototype of body weight support gait training robot. The direct-drive hip and knee joints using servo motor and harmonic reducer are designed for the exoskeleton leg. The follow-up waist support system can move vertically due to employing electric linear actuator, which overcomes the deficiencies of not being back drivable and moving forward and backward existing in similar products. All of these make the training robot more adaptable to individual diversity of patients.2. Key technologies including gait planning and motion control algorithm are studied. Based on the given reference gait trajectory, a new parameter transformation method is proposed to get the gait trajectory automatically for different requirements of stride and step frequency in the rehabilitation process for different patients. Meanwhile, motion control algorithm and harmonically control algorithm of two legs are studied. These algorithms provide theoretical support for the control system of prototype robot.3. Development of control system of prototype robot. Two gait training modes are implemented in the control system,“robot in active”mode and“patient in active”mode. Moreover, real-time training parameters and device status can be displayed. Three levels of prevention are applied to keep safe and reliable operation of both patients and the system.4. The performance testing of the prototype robot has been done. The gait-training interventions were also implemented for two healthy subjects and one paraplegic. The primary results indicate that the functions of the robot prototype have satisfied the design requirements.