Research On Key Techniques Of Perception And Gait Planning Of Lower Extremity Rehabilitation Exoskeleton

With the rapid development of science and technology in the new era,a large number of human-machine interaction robots,represented by rehabilitation exoskeleton,have emerged,which has brought good news to the growing elderly population and the rehabilitation population of lower limb disorders.However,although more and more robots are applied in the field of rehabilitation,which reduces the heavy burden for rehabilitation doctors and provides more scientific rehabilitation training programs for the rehabilitation population,the effect of human-machine interaction of these robots is unsatisfactory,and the perception technology of rehabilitation robots needs to be improved.At the same time,according to the different physiological parameters and rehabilitation stages of the wearers,the rehabilitation robots belonging to the type of human-machine interaction need methods for adaptive gait plannning and trajectory controlling,which is also a difficult problem in this field at present.In order to improve the efficiency of human-machine interaction of the lower extremity rehabilitation exoskeleton,this paper studies the key technologies of the perception and gait planning of the rehabilitation exoskeleton,and builds the suitable upper computer part of the control system of the lower extremity rehabilitation exoskeleton for this subject.For making the robot have the ability to face the complex multi-task environment in the field of practical rehabilitation,the control system of the lower limb rehabilitation exoskeleton,especially the upper computer part,was built and expanded,mainly including the selection and debugging of hardware,the configuration of the real-time environment of the system,the construction of the task scheduling framework and the realization of communication,etc,which provides a safe and efficient control platform for the lower limb rehabilitation exoskeleton robot system and is the basis of the experiment of the prototype and verification for related research technology of the system.In view of the characteristics and difficulties of human-machine interaction in rehabilitation field,this paper proposes a multi-source information perception technology of the lower extremity rehabilitation exoskeleton based on the information of sEMG,inertia,plantar pressure,human-machine interaction,etc.,and studies the collection and application of each information respectively.The application scenario of joint torque estimation based on the SEMG State Equation is mainly aimed at the active participation of the wearer in the rehabilitation training stage,which assists the online control of the robot;the inertial information mainly provides the joint angle information in the stage of gait planning and prototype debugging;a SVM Road Identification Model based on plantar pressure is proposed to meet the needs of the wearer to walking on the actual road in the later stage of rehabilitation;in view of the interaction between the wearer and the exoskeleton,this paper proposes a design method of human-machine interaction sensor verified by experiments,which can provide an important basis for the compliance control of the system during rehabilitation training.The gait planning of the lower extremity rehabilitation exoskeleton robot is based on the rehabilitation training of the wearer and guided by the goal of the gait planning.In this paper,the gait database and gait planning model are established to realize the adaptive planning of the gait trajectory of the lower extremity rehabilitation exoskeleton robot.Based on the construction of the Gait Cycle Dual Support Phase Model,this paper proposes an improved genetic algorithm to find the best gait position for the current wearer,and then obtains the adaptive gait trajectory through the Gait Adjustment Coefficient Model.As the key technology of active rehabilitation,an algorithm for online trajectory control based on Integrated Torque Adaptation Model is proposed.Finally,the prototype experiment is carried out on the platform of the lower limb exoskeleton robot In the early stage,the stability of the control system platform and the accuracy of the gait database are verified on the single leg exoskeleton prototype platform.In the later stage,the practical performance of the task scheduling algorithm and the gait planning are verified on the two legs exoskeleton prototype.