【作者】 吴波；

【导师】 陈照波；

【作者基本信息】 哈尔滨工业大学 ， 机械设计及理论， 2015， 硕士

【摘要】 随着社会的发展,越来越多的人因交通事故、自然灾害、战争或者疾病造成截肢。假肢能够代偿截肢者的部分运动能力,对大腿截肢者回归正常生活、重拾生活信心意义重大。但是目前的假肢产品性能较为一般,很难满足截肢者对假肢的要求。本文就以提高智能假肢行走的步态美观性及其应对复杂运动的能力为目的,提出了使用四连杆膝关节和混合驱动模式的智能假肢结构。并通过理论和仿真研究,验证了智能假肢的步态美观性及其复杂运动下的性能。进行了智能假肢结构的优化。针对步态美观性的要求,选用了四连杆膝关节作为假肢的膝关节结构。建立了四连杆膝关节的和大、小腿参数的优化函数,采用非线性优化方法进行了优化设计。确定了以磁流变阻尼器为主,直流伺服电机为辅的假肢膝关节混合驱动形式,并根据能耗最低原则建立了膝关节力矩的分配方式。借鉴仿人机器人的步态规划和动力学建模方法,建立了智能假肢的运动学模型,并选用劳斯方程建立了其动力学模型。对运动学和动力学模型进行了求解,根据求解结果对假肢步态进行了规划,对磁流变阻尼器和电机等元器件参数进行了选择。搭建了磁流变阻尼器的测试系统,对磁流变阻尼器进行了动力特性测试。采用BP神经网络和ANFIS技术分别建立了磁流变阻尼器的正向模型和逆向控制器模型,并根据测试数据,对磁流变阻尼器的正向和逆向模型进行了学习训练。选择性能最佳的磁流变阻尼器正向模型和控制器模型建立了磁流变阻尼器系统模型,并对系统模型的精确性进行了验证。搭建了智能假肢的虚拟样机联合仿真平台,对智能假肢在平地行走、上楼梯、下楼梯、下斜坡等不同运动状态下的运动进行了仿真。仿真结果表明,运用了磁流变阻尼器的动力型智能假肢在平地行走时具有很高的步态美观性,并且能够完成上楼梯、下楼梯和下斜坡这几个复杂的运动,具有很高的路况适应性。更多还原

【Abstract】 With the development of the society, more and more people are amputated because of traffic accidents, natural disasters, battles or diseases. Prostheses can partially compensate for some lost athletic abilities, which is significant to the amputees to return to normal life and regain confidence in life. However, the performance of current prosthetic products is still more general and can hardly meet the demands of amputees. With the goals of improving the aesthetic property of walking gait and the capacity to cope with complex road environments, four-bar linkage knee and the hybrid-driven mode are employed to design the intelligent prosthesis. Finally the required performance is verified by theory research and simulation.The optimized design of the intelligent prosthesis is conducted. Four-bar linkage knee is chosen for prosthetic knee joint structure to get a better gait. The optimal objective functions of four-bar linkage knee, thigh and calf are established and solved with nonlinear optimization method. The prosthetic driven-mode relied mainly on MR damper while DC motor subsidiary is determined. And the knee-torque distribution method is setted up based on the principle of energy consumption minimization.Reference to the gait planning and dynamic modeling methods of humanoid robots, the kinematics model of the intelligent prosthesis is established and the dynamic model is established using Routh equation. The models are solved based on the walking date. The gait of prosthetic movement is planed and the component’s parameter is selected based on the results.The test system is constructed to conduct the MR damper dynamical test. Forward models and inverse models of the MR damper are established by using BP neural network and ANFIS. The parameter training and learning of the MR damper models are conducted based on the test data. The MR damper system model is conducted form the selected models with best integrated performance. And the accuracy of the MR damper system model is verified by simulation.The virtual prototype and united simulation platform for the intelligent prosthesis is established. The motion simulation is run under complex road environments such as walking, up stairs, down stairs and down a ramp. The results indicate that active intelligent prosthesis used MR damper has an aesthetic property of gait when walking and a strong adaptability to different road situations.更多还原