康复运载型下肢外骨骼的步态规划与轨迹跟踪控制研究

发布时间：2017-12-28 02:25

本文关键词：康复运载型下肢外骨骼的步态规划与轨迹跟踪控制研究　出处：《南京理工大学》2017年硕士论文　论文类型：学位论文

【摘要】：可穿戴式外骨骼将机器人提供的强大机械能量与人的智能相结合,是一种新型机电一体化装置,能够提供身体支撑、运动辅助、机能增强等功能。在当今人口老龄化趋势日益严重、单兵作战能力亟待提高的形势下,外骨骼已经成为机器人领域的研究热点,并可以应用到众多领域。本文以康复运载型下肢外骨骼系统为对象开展设计研究。在人体下肢生理结构与步行分析的基础上,文中给出了具有10自由度的系统结构设计,根据康复运载这一侧重点进行了稳定步态规划与轨迹跟踪控制研究。主要研究工作如下:1)下肢外骨骼系统结构设计。归类总结分析现有研究成果,根据人体步行特性确定系统功能实现所需的自由度数目及其配置,完成系统的结构设计。2)支撑腿与摆动腿运动学、动力学模型的建立。由步行分析得出下肢外骨骼的混杂系统特性,从而分别对支撑腿和摆动腿进行运动学和动力学分析。运动学分析中采用改进的D-H方法,包含了正、逆运动学的求解验证。在此基础上,通过拉格朗日方程法得出了动力学模型。结合切换系统描述的离散事件动态特性,建立完整的外骨骼系统混杂模型。为模拟真实环境并简化混杂系统数学模型的构建以及后续控制器的设计,采用动力学仿真软件ADAMS建立了下肢外骨骼系统的虚拟样机。3)稳定步态的规划。运载型这一功能特点表明外骨骼占据主导地位并带动穿戴者实现正常步行,使得稳定步态规划成为重要研究内容。对系统步行稳定的分析采用的是零力矩点(ZMP)理论,即稳定步行需保证ZMP始终位于支撑域内。根据人体步行周期的分析,确定了步态规划的关键时刻约束条件,通过三次样条插值获得参数化的连续轨迹。结合稳定裕度计算,设计了遗传算法优化步态参数所需的目标函数。获得的稳定连续步态经逆运动学可得出预期关节轨迹,并通过ADAMS动力学仿真验证其稳定性。4)轨迹跟踪控制的研究。稳定步行的最终实现需要各关节自由度对预期轨迹进行精确跟踪。基于动力学分析所得系统模型的数学描述设计了计算力矩加PD反馈控制器,针对无法获得精确数学表达的虚拟样机被控对象设计了无模型思想的TDE-iPID与TDE-iPIDESMC控制器。TDE-iPID采用时延估计来获得系统的极局部模型,通过iPID反馈形成控制回路。在TDE-iPID基础上建立时延估计误差模型,利用等效滑模控制(ESMC)进行估计误差补偿。通过MATLAB/Simulink与ADAMS的联合仿真分析验证了 TDE-iPID轨迹跟踪控制策略的合理有效性,以及ESMC对跟踪结果的改善。
[Abstract]:Wearable exoskeleton, which combines powerful mechanical energy and human intelligence, is a new electromechanical integration device, which can provide body support, motion assistance, function enhancement and other functions. Exoskeleton has become a research hotspot in the field of robotics, and it can be applied to many fields when the trend of population aging is becoming more and more serious. In this paper, a study was conducted on the rehabilitation of the exoskeleton system of the lower extremities. Based on the analysis of the physiological structure and walking of human lower limbs, a system structure design with 10 degrees of freedom is given. Based on the emphasis of rehabilitation, the stable gait planning and trajectory tracking control are studied. The main research work is as follows: 1) the structure design of the exoskeleton system of the lower extremities. The existing research results are classified, analyzed, and the functions of the system are determined according to the walking characteristics of the human body to achieve the required degrees of freedom and their configuration, and the structural design of the system is completed. 2) the establishment of the kinematics and dynamics model of the supporting leg and the swinging leg. The characteristics of the hybrid system of the exoskeleton of the lower extremities were obtained by walking analysis, and the kinematic and dynamic analysis of the support leg and the swinging leg were carried out respectively. The improved D-H method is used in the kinematic analysis, which includes the validation of the positive and inverse kinematics. On this basis, the dynamic model is obtained by the Lagrange equation method. Combined with the dynamic characteristics of discrete events described by the switching system, a complete hybrid model of the exoskeleton system is established. In order to simulate the real environment and simplify the construction of the mathematical model of hybrid system and the design of subsequent controller, a virtual prototype of the exoskeleton system of lower extremities was established by using the dynamic simulation software ADAMS. 3) the planning of stable gait. The features of the carrier type show that the exoskeleton occupies the dominant position and drives the wearer to walk normally, making the stable gait planning an important research content. The analysis of the stability of the system adopts the zero moment point (ZMP) theory, that is, the stable walking needs to ensure that the ZMP is always in the supporting domain. According to the analysis of human walking cycle, the key constraint conditions of gait planning are determined, and the parameterized continuous trajectories are obtained by three spline interpolation. Combined with the calculation of stability margin, the objective function required by genetic algorithm to optimize gait parameters is designed. The desired joint trajectory can be obtained by the inverse kinematics of the stable continuous gait, and the stability is verified by ADAMS dynamic simulation. 4) the research of trajectory tracking control. The final realization of stable walking requires the accuracy of the degree of freedom of each joint to track the expected trajectory accurately. Based on the mathematical description of the system model based on the dynamic analysis, the computed torque plus PD feedback controller is designed. The TDE-iPID and TDE-iPIDESMC controller without model thinking is designed for the object controlled by the virtual prototype which can not get the exact mathematical expression. TDE-iPID uses the time delay estimation to obtain the extremely local model of the system, and forms the control loop through the feedback of iPID. The time delay estimation error model is established on the basis of TDE-iPID, and the estimation error compensation is made by using the equivalent sliding mode control (ESMC). The joint simulation analysis of MATLAB/Simulink and ADAMS validates the rationality and effectiveness of the TDE-iPID trajectory tracking control strategy and the improvement of the tracking results by ESMC.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242

【相似文献】