Dynamic Design Optimization Under Uncertainty For The Lower Extremity Exoskeleton

Lower extremity exoskeleton(LEEX)robot is a wearable man-machine mechanical system,which can enhance or restore the athletic performance of the human body.Due to enhancing the walking speed and weight bearing capacity of normal population,LEEX has been applied to special operations,such as the individual soldier combat,polar expedition,earthquake rescue,and so on.It has been also employed to provide walking assistance and rehabilitation training for people with lower limb dysfunction.In view of the broad application prospects of LEEX in the fields of the medical rehabilitation,military field,space exploration,elderly and handicapped people assistance,a great deal of in-depth and imperative investigations have been done for LEEX.Because LEEX is a typical man-machine coupling system,the comfort,and safety of the wearer are important indicators in the wear performance of LEEX.Moreover,the energy saving and lightweighting has also become an important trend in the development of LEEX.Therefore,the study area of the safety,comfort and energy saving is of great practical significance.Uncertainty is ubiquitous in exoskeleton systems during the systems’ lifecycle,which will affect both comfort and safety of the wearer,and may even cause harm because the motion of exoskeleton systems changes obviously under uncertainty so that human body is collided by an exoskeleton.This dissertation has been focused on the dynamic uncertainty design of LEEX in design stage for safety,comfort and energy saving in its operation.The contributions of this dissertation are summarized as follows:(1)Development of structure design model for LEEX based on the movement characteristics of the lower limb.In order to achieve LEEX’s design target of nimble mechanism,highly consistent man-machine coupling system and safely wearing.The motion trajectory functions,kinematic model and corresponding algorithm of joint angles in lower limb have been established through well-studied motion characteristics.Based on incomplete matching semi-anthropomorphic design,the reasonably matching degrees of freedom and selecting the actuated joints for LEEX,the virtual model of LEEX has been built.This model lays the engineering foundation for dynamic uncertainty design of LEEX.(2)Development of the man-machine coupling consistency model for LEEX with internal uncertainties and two time-variant reliability analysis methodsTo measure the safety of LEEX under internal uncertainties,the man-machine coupling consistency has been defined in the light of the degree of coupling between practical movement locus of LEEX and theoretical movement locus of the wearer.Moreover,the law of man-machine collision has been revealed.The mutually independent consistency model and the dependent consistency model have been built by tracking the source and transmittal mode of internal uncertainties,considering the dependency of all joints.Due to the peculiarity of those two models,two time-variant reliability analysis methods have been proposed to reveal the evolvement rule of LEEX’s safety under internal uncertainties.As the gait cycle goes on,gradual accumulation of inconsistency reduces the security of LEEX,which provides the theoretical basis for energy conservation,security and comfort design.(3)Development of the man-machine coupling consistency model for LEEX under a stochastic terrain and a time-variant reliability with multiple failure modes and multiple temporal parameters analysis methodAccording to the source of uncertainties of LEEX,uncertainties are divided into internal uncertainties and external uncertainties.External uncertainties of LEEX mainly comes from working condition,namely the randomness from the stochastic terrain.Considering the effect of the stochastic terrain for the end trajectory of LEEX,the difference between the actual end trajectory under uncertainty and the theoretical end trajectory has been measured based on direct kinematics.Moreover,consistency models of the stance phase and swing phase have been established separately.The time-variant reliability analysis with multiple failure modes and multiple temporal parameters is proposed to reveal the evolvement rule of LEEX’s safety under the stochastic terrain,which offers a theoretical basis for security and comfort design.(4)Development of the consistency energy conservation model for LEEX and an efficient decoupling algorithm for time-variant reliability-based design optimization(RBDO)According to the configuration scheme of the actuated joints and the actuator selected by the virtual prototype of LEEX,the hydraulic drive with the largest energy consumption is selected as the design object.Then energy consumption models of the hydraulic cylinders have been established under the condition of ensuring the normal operation of the LEEX.In order to ensure the safety of LEEX,the consistency energy conservation model of LEEX have been established by constructing human-machine coupling consistency constraints.Meanwhile,consider the properties of consistency energy conservation model,an efficient time-variant RBDO decoupling algorithm has been established to solve the optimal allocation strategy of hydraulic cylinders.This timevariant RBDO decoupling algorithm expands the calculation method of dynamic uncertainty design.Compared with determined energy conservation optimization,the consistency energy conservation model has shown significant advantages for ensuring the safety of LEEX.(5)Development of a unified framework for safety and comfort design of the LEEX and a time-variant reliability-based robust optimization design algorithmSafety and comfort are two main attributes of dynamic uncertainties design for LEEX.In the uncertainties design,safety and comfort should be satisfied at the same time.Tracking the transmission of uncertainties in the LEEX,according to the connotation of robustness,the comfort of LEEX is characterized by the degree of fluctuation of the joints under uncertainty,and a robust performance function for characterizing comfort has been established.Meanwhile,based on the definition of consistency,the safety of LEEX is described by the difference between the actual value and the theoretical value of the joints under uncertainty,and a consistent performance function for characterizing safety has been established.A unified model of the consistency-based robust optimization design model for LEEX has been established,which truly realized the combination of safety design and comfort design,and formed a dynamic uncertainty design system for LEEX.The consistency-based robust model is characterized by time-variant reliability-based robust design with multiple failure modes and multiple temporal parameters.Therefore,a solution algorithm is proposed to perfect the theoretical system and solution method of dynamic design optimization under uncertainty.