Research On Human-robot Motion Integration Mechanism And Evaluation Method Of Lower Limb Exoskeleton Robot

Power walking lower extremity exoskeleton robot is an intelligent equipment which can provide help for the elderly to walk and improve the walking ability of people with lower extremity motion disorder.With the increase of age,the function of the human body gradually degrades.More and more elderly people or people with lower extremity motor disorders have a strong demand for wearable intelligent booster devices.Lower limb exoskeleton robot introduces human intelligent decision-making ability,and improves the human-robot motion integration.As a wearable auxiliary equipment,people have higher and higher requirements for its safety and human-robot interaction.The integration of human-robot motion is a hot and difficult point in the research of exoskeleton robot.The focus of its research at home and abroad is mainly to improve the human-robot motion integration by optimizing the motion flexibility of exoskeleton body and the stability of control strategy,but there are relatively few studies on whether exoskeleton equipment can follow the cooperative movement of human body when wearing exoskeleton,and the wearing comfort in the fusion of human-robot motion.Therefore,it is of great significance to study the wearing comfort of exoskeleton robot in improving the intelligent level and application effect of exoskeleton robot.External skeletal walker in the process of human motion communion problem,this thesis takes anhui university of engineering exoskeleton power robot as the research object,using the existing laboratory equipment for foot pressure and methods of kinematics and physiology,such as electric information data,by means of multimodal information fusion mathematical index evaluating the comfort of wearable and comfortable quantitative indicators is established at the same time.The research contents are as follows:(1)The construction of the experimental platform of vital information collection based on the mechanism of human-robot interaction.Based on the theory of human biology,the joint structure of human lower limb is simplified as a linkage mechanism,the gait phase is divided,the movement characteristics are analyzed,the exoskeleton structure is optimized,the joint drive is matched,and the effect of human-robot movement is optimized.The mechanism of physiological and kinematic signals generation is analyzed,and a multi-sensor information acquisition platform is constructed based on the human motion mechanism and the existing sensor equipment in the laboratory,it provides a data base for the research of human-robot motion integration evaluation method.(2)Aiming at the problems of large sample set data of physiology and kinematics information,difficult to extract characteristic values and unclear motion state recognition,the thesis proposes a method of motion state recognition based on the characteristics of physiology and kinematics respectively.The physiological and kinematic information is denoised,eigenvalue extraction,mathematical modeling and other processing,matching the appropriate algorithm for motion state recognition research.(3)Aiming at the problem of low recognition accuracy of single sensor,a human-robot motion state recognition method based on vital multi-modal information fusion is proposed in this thesis.The information of physiology and kinematics is fused,and the fused mathematical model is established to do the fused motion recognition.Through comparative analysis,the accuracy rate of motion state recognition after fusion reaches 94.7%,which is better than the motion state recognition result using a single sensor,and lays a foundation for the subsequent study of human-machine motion fusion of exoskeleton.(4)Experimental study on human-robot collaboration for pedestrian assistance.A complete human-robot motion integration evaluation method was established from both subjective and objective aspects,and an experimental scheme of human-robot coordination assisted walking was designed.To conduct an experimental study,and the experimental data were processed and analyzed.The results show that method is effective.In response to the questions raised,in this thesis,the mechanism of human movement of blending is studied,optimize the structure design of ontology and joint drive,analysis and establish the experimental platform for physiology and kinematics,the human-robot movement compatibility evaluation mathematics model is established,to wear an exoskeleton walker walking experiment,verify the validity of the evaluation method,It lays a foundation for subsequent research on human-robot motion integration of exoskeleton and effectively promotes the improvement of intelligence level of lower limb exoskeleton robot.