In view of the current testing and evaluation methods of lower limb power exoskeleton man-machine effectiveness are relatively simple and lack of unified standards,resulting in onesidedness and fuzziness of man-machine effectiveness analysis of lower limb power exoskeleton.Based on the basic theory of human lower extremity biolysis planing,the basic theory of human lower extremity kinematics and the basic research related to the humanmachine effectiveness analysis of lower extremity power exoskeleton.From the two dimensions of subjective human feeling and objective physiological feeling,the QUEST level judgment model of lower limb movement comfort level was established to realize the effective monitoring and analysis of the comprehensive human feeling in the process of lower limb movement.On this basis,relevant indexes of human-computer interaction efficiency were introduced to analyze the human-computer effectiveness of lower extremity power assisted exoskeleton,and a method of human-computer effectiveness analysis of lower extremity power assisted exoskeleton based on fuzzy Kohonen clustering network was proposed.Finally,in order to improve the efficiency and universality of human-machine analysis of lower limb power exoskeleton,this study builds a human-machine effectiveness analysis system of lower limb power exoskeleton based on the above research results,and verifies the reliability and effectiveness of the system through case optimization.This paper mainly carries out the following work:(1)Related basic theory research: Firstly,the importance of a more effective and comprehensive man-machine effectiveness test and evaluation method for the design and optimization of lower limb power exoskeleton is discussed,and the research direction and breakthrough point of this paper are clarified by summarizing the research status at home and abroad in this field.The theory of lower limb biodissociation planing and lower limb kinematics is elaborated.The man-machine effectiveness evaluation methods and related classification and clustering algorithms of lower limb power exoskeleton are described.The reliability and effectiveness of the system are verified by an example optimization.(2)QUEST grade judgment model for lower limb exercise comfort level: Aiming at the problems of fuzziness and one-sidedness in the evaluation of comfort feeling in the current lower limb exercise process,a Quest-based analysis and judgment model for lower limb exercise comfort level was proposed,which integrated subjective and objective comfort feelings.Through the design of non-weight-bearing lower limb squat test,the original data related to human feelings were obtained,and the calculated data and K-means ++ clustering results were combined to form the lower limb exercise comfort level determination data set.Through training and iteration,a Quest-based lower limb exercise comfort level analysis and determination model was established.The experimental results show that this model can effectively determine and classify the level of comfort level in the process of lower limb movement,which to a large extent makes up for the fuzzy evaluation of comfort feeling of a single comfort index,and also provides theoretical support for the subsequent clustering analysis method of lower limb power exoskeleton man-machine effectiveness proposed by fuzzy Kohonen clustering network.(3)Human-machine effectiveness analysis method of lower extremity power exoskeleton based on fuzzy Kohonen clustering network: There is a lack of unified standards for humanmachine effectiveness analysis of lower extremity power exoskeleton at present,and there are certain problems of one-sidedness and fuziness in man-machine effectiveness analysis of lower extremity power exoskeleton based on a single evaluation standard of human perception or mechanical effectiveness.In this chapter,the effectiveness evaluation system of lower limb power exoskeleton human-machine system is established from the two dimensions of human sensory evaluation and human-machine interaction efficiency.The multi-task assisted exoskeleton gait experiment was designed to obtain the relevant data of human perception and human-computer interaction efficiency.The original data were extracted and calculated by relevant algorithms,and the effectiveness evaluation data set of the lower extremity assisted exoskeleton human-computer system was constructed.On this basis,the fuzzy Kohonen clustering network was used to cluster and analyze the data set,and the human-machine effectiveness analysis of the lower limb power exoskeleton was realized based on the fuzzy Kohonen clustering network.Finally,compared with the clustering results of a single evaluation index,the advantages of the proposed method are verified.Experiments show that this method can effectively make up for the limitation of man-machine effectiveness test and evaluation method of lower extremity power exoskeleton caused by single evaluation index.It can provide more reliable and effective theoretical support for the design and optimization of exoskeleton equipment.(4)Human-machine effectiveness analysis system for lower extremity power exoskeleton:Aiming at the low efficiency of testing and evaluating the human-machine effectiveness of lower limb power exoskeleton and the one-sidedness and fuzziness caused by the analysis of a single type of index,based on the QUEST level judgment model of lower limb motion comfort level and the clustering analysis method of human-machine effectiveness of lower limb power exoskeleton based on fuzzy Kohonen clustering network,The manmachine efficiency analysis system of lower extremity power exoskeleton was built.The reliability and effectiveness of the system are verified by an optimization example of lower extremity exoskeleton.The results show that the optimization effect of the lower limb power exoskeleton prototype is obvious according to the man-machine efficiency analysis results of the output system.It can provide more efficient analytical means and reliable theoretical reference for the design and optimization of lower limb power exoskeleton. |