Gait Planning And Adaptive Control Of Lower Extremity Exoskeleton Robots

The lower extremity exoskeleton robots are a class of wearable robots which provide the melding of people and robots,particularly,the enhancement of the human joint functions.The application demands for lower extremity exoskeleton robots are increasing in the military,industrial and medical fields.To improve the performance of the lower extremity exoskeleton robots,researchers have studied the related techniques of lower extremity exoskeleton robots.Therein,the gait planning and control of the lower extremity exoskeleton robots are the popular researching fields.In this thesis,a lower extremity exoskeleton robot with 4 degrees of freedom per leg is designed,where three degrees of freedom in the hip and knee joints are active and one degree of freedom in the ankle joint is passive.The gait planning and control strategies of the lower limb exoskeleton robot are studied in this thesis.A stable walking trajectory is planned and a stable adaptive controller is designed to satisfy the requirements for the lower extremity exoskeleton robot.The main research work of this thesis is summarized as follows:(1)A set of lower extremity exoskeleton robot system is designed according to the requirement of the walking movement.The special structure of the system can provide the anthropomorphic design and the adjustable size to match different people.(2)Addressing the gait planning of lower extremity exoskeleton robot,a method of gait planning combining ZMP stability criterion and polynomial interpolation algorithm is proposed.The method refers to the basic characteristics of the human gait,realizes the parametric representation of the gait,and simplifies the representation of the gait.(3)A fuzzy adaptive controller combining virtual tunnels is designed to control the lower extremity exoskeleton robot.The controller combining the fuzzy approximator and the iterative learning method can deal with the uncertainty of the human-robot dynamics model and the periodic external disturbance effectively.By constructing virtual tunnels in the joint space,the subject is allowed to move freely in the virtual tunnels and affects the movement of the lower extremity exoskeleton robot.The active participation discussed in the thesis can solve the problem of insufficient participation for the subjects in the fixed trajectory tracking.