【作者】 李东；

【导师】 李飞；

【作者基本信息】 沈阳工业大学 ， 工程硕士（专业学位）， 2021， 硕士

【摘要】 下肢外骨骼机器人是一种穿戴在人体外部的智能机械装置,结合生物医学、机械制造原理、控制理论和计算机技术等学科的知识,构成了一个复杂的智能系统。区别于传统的机器人,其不仅要作为机器单独工作,还要与人作为一个整体协同运动,通过控制机器人的运动为穿戴者提供预定义方案的康复训练、支撑、保护等功能。本文为下肢行动有障碍者设计了康复型外骨骼机器人,包括其机械结构设计、运动学和动力学模型的建立、人机协调控制方法的研究,并通过虚拟样机联合控制仿真验证了控制方案的有效性。本文主要研究内容如下:分析研究了人体下肢生理结构及运动机理,设计出一款自由度贴合人类正常行走规律、便于穿戴的下肢外骨骼的机械结构,并利用Solidworks完成了其三维模型的建立。分析外骨骼运动学和动力学。利用D-H法建立外骨骼机构正运动学模型,分析外骨骼机器人关节角度的变化对位姿的影响;利用解析法建立外骨骼机构逆运动学模型,分析外骨骼机器人的位姿变化对关节角度的影响,通过MATLAB对运动学模型进行仿真分析,验证了其正确性。利用拉格朗日法建立外骨骼机构动力学模型,分析运动状态下关节力矩与关节角度、角速度之间的关系。通过ADAMS软件对外骨骼机构进行动力学建模,添加摩擦力、重力、人机交互接触力等,对运动时各关节的受力情况进行仿真分析,验证了外骨骼机器人动力学模型的正确性,为人机协调控制研究奠定了基础。分析比较几种常见的机器人控制方法,针对经典PID控制策略参数不易整定,抗干扰能力差的缺点,研究了一种基于模糊规则改进的PID控制方法。利用模糊规则对PID控制器的参数实时调整,能够增强控制系统的鲁棒性。针对下肢外骨骼机器人这样一个多输入、多输出的非线性复杂控制系统,研究了滑模控制方法,其对机器人系统参数的变化及扰动不灵敏,有很好的鲁棒性。利用ADAMS与MATLAB/Simulink进行虚拟样机联合控制仿真,对外骨骼机构的调节能力、响应速度、跟踪效果进行了详细分析,验证了基于模糊规则的PID控制方案和滑模控制方案的有效性。更多还原

【Abstract】 The lower limb exoskeleton robot is an intelligent mechanical device worn outside the human body.It combines the knowledge of biomedicine,mechanical manufacturing principle,control theory and computer technology to form a complex intelligent system.Different from the traditional robot,it not only works as a machine,but also cooperates with human as a whole.By controlling the movement of the robot,it provides the wearer with pre-defined rehabilitation training,support,protection and other functions.In this paper,a rehabilitative exoskeleton robot is designed for the patients with lower extremity dyskinesia,including the design of its mechanical structure,the establishment of kinematics and dynamics model,and the research of man-robot coordinated control method.The effectiveness of the control scheme is verified by the virtual prototype joint control simulation.The main contents of this paper are as follows:The physiological structure and movement mechanism of human lower limbs were analyzed.This paper designs a mechanical structure of lower limb exoskeleton with degree of freedom fitting human normal walking law and easy to wear,and uses Solid Works to complete the establishment of its three-dimensional model.The kinematics and dynamics of exoskeleton are analyzed.The D-H method is used to establish the forward kinematics model of exoskeleton mechanism and analyze the influence of the change of exoskeleton robot joint angle on the posture the analytical method is used to establish the inverse kinematics model of exoskeleton mechanism and analyze the influence of the change of exoskeleton robot joint angle.The kinematics model is simulated and analyzed by MATLAB to verify its correctness.The dynamic model of exoskeleton mechanism is established by Lagrange method,and the relationship among joint torque,joint angle and angular velocity is analyzed.The dynamics model of exoskeleton mechanism is established by ADAMS software.The friction force,gravity force and human-robot interaction force are added to simulate and analyze the force of each joint.The correctness of the dynamics model of exoskeleton robot is verified,which lays a foundation for the research of human-robot coordination control.Several common robot control methods are analyzed and compared.Aiming at the shortcomings of the classical PID control strategy,such as the difficulty of parameter tuning and poor anti-interference ability,an improved PID control method based on fuzzy rules is reseached.Using fuzzy rules to adjust the parameters of PID controller in real time can enhance the robustness of the control system.A sliding mode control method is reseached for the multi input and multi output nonlinear complex control system of lower limb exoskeleton robot,which is insensitive to the changes and disturbances of robot system parameters and has good robustness.Adams and MATLAB / Simulink are used to simulate the joint control of virtual prototype,and the regulation ability,response speed and tracking effect of exoskeleton mechanism are analyzed in detail.The effectiveness of PID control scheme based on fuzzy rules and sliding mode control scheme is verified.更多还原