CN103230312A - Optimization method for three hinge points of power-assisted exoskeleton knee - Google Patents

Abstract

The invention discloses an optimization method for three hinge points of a power-assisted exoskeleton knee and subjects power-assisted exoskeleton knee parameters to optimizing processing. The optimization method for three hinge points of the power-assisted exoskeleton knee comprises establishing a three-hinge point mathematical model of a thigh, a crus and a hydraulic cylinder; utilizing location parameters of the three hinge points as optimization parameters; utilizing the least stress of the hydraulic cylinder as an optimization objective; utilizing corresponding human body sizes as constraint conditions of the model; and obtaining the location parameters of all the hinge points through an improved particle swarm optimization algorithm so as to provide parameters for modeling and dynamic analyses of exoskeleton and the dynamical system of the exoskeleton.

Description

A kind of assistance type ectoskeleton knee joint tri-joint mechanism optimization method

Technical field

The present invention relates to optimize the design of algorithm design and frame for movement, relate in particular to a kind of for assistance type ectoskeleton knee joint tri-joint mechanism optimal design based on the particle swarm optimization algorithm design of simulated annealing and the mathematical modeling of ectoskeleton knee joint tri-joint mechanism, and by calculating final ectoskeleton knee joint tri-joint mechanism location parameter.

Background technology

Human body ectoskeleton technology derives from the magical ability of nature biotechnology, be to simulate the shell of Crustacean, constitute according to human muscle, skeleton and joint, utilize modern science and technology for human body is equipped with corresponding ectoskeleton, be mainly the human body weight loading exercise and provide support and assist, improve the heavy burden ability of human body and the continuous capability of weight loading exercise.

According to human body heavy burden principle, the human body ectoskeleton mainly comprises AIR CONVECTOR TECH and lower limb carrying boosting mechanism.AIR CONVECTOR TECH is used for laying human body and bears a heavy burden, bear a heavy burden and mainly be passed to ground by lower limb carrying boosting mechanism, thereby boosting mechanism is the core of whole human body ectoskeleton frame for movement, is made up of the thigh, shank, hydraulic cylinder and the corresponding joint that are complementary with skeleton.Wherein hydraulic cylinder mainly provides the required potential energy of vertical load sink-float motion in the man-machine portable motor process, and adjusts according to the variation of human body different motion attitude.Therefore, the parameter of choose reasonable lower limb carrying boosting mechanism has directly determined the performance of whole system, needs to use the reasonable optimizing algorithm that it is carried out to optimize and analyzes, and obtains the optimal solution of tri-joint mechanism parameter.

The carrying situation of human body ectoskeleton lower limb boosting mechanism is complicated and changeable, and the tri-joint mechanism structure that constitutes with thigh, shank and hydraulic cylinder is typical nonlinear restriction problem; And particle swarm optimization is a kind of bionic intelligence optimization algorithm of simulating birds colony foraging behavior, and it can avoid tradition to optimize the shortcoming that algorithm relies on the problem characteristic, is applicable to the Optimal Structure Designing problem of complicated multivariable nonlinearity object function.Therefore on standard particle swarm optimization basis, using the improvement particle swarm optimization based on simulated annealing, ectoskeleton tri-joint mechanism stress model is carried out structural dimension optimization, is the steps necessary of ectoskeleton research.

In the research of present ectoskeleton frame for movement, emphasis concentrates on aspects such as the integrally-built intensity of ectoskeleton, stiffness analysis, the emphasis of research is parts such as ectoskeletal thigh and shank, and it is whether reasonable for the hydraulic cylinder in the dynamical system and installation site thereof, but do not carry out corresponding research, also thigh, shank and hydraulic cylinder are not joined together to carry out mathematical modeling, thereby cause the stressed of each part of ectoskeleton and hydraulic cylinder reasonably not distribute, over-burden to cause dynamical system.

Summary of the invention

In view of the above deficiency of prior art, the present invention provides a kind of assistance type ectoskeleton knee joint tri-joint mechanism optimization method, makes it to be applied to the optimization of knee joint tri-joint mechanism of ectoskeleton structural design to overcome the above deficiency of prior art.

The technical solution adopted in the present invention is:

A kind of assistance type ectoskeleton knee joint tri-joint mechanism optimization method is optimized processing to assistance type ectoskeleton knee joint parameter, comprises following steps:

1) set up the position relation of joint ectoskeleton knee joint tri-joint mechanism: tri-joint mechanism comprises knee joint O ₂₂, the last hinge O that is connected with thigh of hydraulic cylinder ₂₁The following hinge O that is connected with shank with hydraulic cylinder ₂₃, the distance between last hinge and the knee joint is the long S of bar ₁, the distance between following hinge and the knee joint is the long S of bar ₃, the distance between the upper and lower hinge is the long S of bar ₂

2) by the improvement micropartical colony optimization algorithm based on simulated annealing, be optimization aim with the loaded hydraulic cylinder minimum, finally obtain the long S of bar ₁, the long S of bar ₂With the long S of bar ₃Value, and this value imported follow-up manufacturing process controlled.

The inventive method is the rational installation site that obtains hydraulic cylinder, sets up the tri-joint mechanism mathematical model of thigh, shank and hydraulic cylinder, is optimization aim with the carrying minimum of hydraulic cylinder, establishes corresponding body dimension and is the restriction range of model; For obtaining reliable optimization solution, utilize the renewal of inertia weight that the standard particle swarm optimization is improved, adopt the mode that dynamically reduces inertia weight ω that it is upgraded along with the variation of iterations, thereby accelerate convergence of algorithm speed; By being the parameter assignment in the algorithm, thereby obtain three location parameters between the hinge, i.e. the rational installation site of hydraulic cylinder.

The new ectoskeletal knee joint tri-joint mechanism of the assistance type optimization method that is used for that adopts that the present invention proposes, by setting up the tri-joint mechanism mathematical model of thigh, shank and hydraulic cylinder, location parameter with three hinges is parameters optimization, stressed minimum with hydraulic cylinder is optimization aim, with the constraints of corresponding body dimension as model, by improved particle swarm optimization algorithm, obtain the location parameter of each hinge, thereby provide parameter for modeling and the dynamic analysis of ectoskeleton and dynamical system thereof.By this optimization method that is applied to the knee joint tri-joint mechanism of assistance type ectoskeleton structural design, can obtain the installation site parameter of rational hydraulic cylinder, for parameter configuration and the light-weight design of dynamical system provides foundation, and can provide modeling parameters for the integrally-built motion analysis of ectoskeleton and dynamics simulation.

Description of drawings

Fig. 1 is ectoskeleton knee joint tri-joint mechanism mechanism model figure.

The specific embodiment

Below in conjunction with accompanying drawing, the present invention is described in further detail.

As shown in Figure 1, the ectoskeleton knee joint tri-joint mechanism comprises knee joint O ₂₂, the last hinge O that is connected with thigh of hydraulic cylinder ₂₁The following hinge O that is connected with shank with hydraulic cylinder ₂₃, the distance between last hinge and the knee joint is S ₁, the distance between following hinge and the knee joint is S ₃, the distance between the upper and lower hinge is S ₂It is balance sysmte that tri-joint mechanism mechanism is one, it is made a concerted effort is 0, and with the stressed power F that is reduced to a vertical direction of thigh upper extreme point, reasonably duty is to bear a heavy burden mainly to be born by thigh and shank, hydraulic cylinder provides antidumping moment, the bearing capacity F that hydraulic cylinder is desirable _CylinderBe 0, can obtain each variable and S according to the cosine law ₁, S ₂And S ₃Functional relationship;

When ectoskeleton was positioned at erectility, it was balance sysmte that tri-joint mechanism mechanism is one, and it is made a concerted effort is 0, and the stressed power F that is reduced to a vertical direction with the thigh upper extreme point carries out force analysis to entire mechanism, obtains F _CylinderRelational expression with F:

When man-machine portable was upright, the most rational duty was to bear a heavy burden mainly to be born by thigh and shank, and hydraulic cylinder provides antidumping moment, and the desirable bearing capacity F of hydraulic cylinder is 0, and therefore, the object function of ectoskeleton knee joint tri-joint mechanism structure parameter optimizing is following formula.

Can obtain the relation of each variable and the long S1 of bar, S2 and S3 in the object function according to the cosine law:

${\mathrm{\&beta;}}_1=\arccos \frac{S_1^2+{\mathrm{<figure-callout\; id="2"\; label="S"\; filenames="HDA00003041181500011.png"\; state="\{\{state\}\}">S</figure-callout>}}_2^2-S_3^2}{2S_1{\mathrm{<figure-callout\; id="2"\; label="S"\; filenames="HDA00003041181500011.png"\; state="\{\{state\}\}">S</figure-callout>}}_2},$ ${\mathrm{\&beta;}}_2=\arccos \frac{S_1^2+S_3^2-{\mathrm{<figure-callout\; id="2"\; label="S"\; filenames="HDA00003041181500011.png"\; state="\{\{state\}\}">S</figure-callout>}}_2^2}{2S_1{S}_3}$

The maximum heavy burden of normal adult is 60kg, add ectoskeleton deadweight 20kg, so power F maximum is got 800N.According to the standard GB/T10000-1988 of China adult physical dimension, adopt the human body standard of southern china people the 95th percentile, the restriction range of setting up design variable is:

In order to strengthen the optimizing ability of particle swarm optimization algorithm, utilize the renewal of inertia weight that the standard particle swarm optimization is improved; Inertia weight ω is as influencing the most important parameter that particle speed upgrades, directly show the control to velocity magnitude and direction, bigger ω is conducive to improve the global optimizing ability of algorithm, less ω then can accelerate convergence of algorithm speed, adopts the mode that dynamically reduces inertia weight ω that it is upgraded along with the variation of iterations:

$\mathrm{\&omega;}={\mathrm{\&omega;}}_{\max }-\frac{t\&times;({\mathrm{\&omega;}}_{\max }-{\mathrm{\&omega;}}_{\min })}{t_{\max }}$

ω in the formula _MaxAnd ω _MinBe maximum and the minima of inertia weight ω, t is iterations, t _MaxIterations for maximum.After the solution of a new generation was calculated, the adaptive value that it is corresponding with the solution of previous generation compared, if new value is then accepted in variation delta E≤0, otherwise according to exp(-Δ E/T) the new value of rand (0,1) criterion acceptance.

According to the standard GB/T10000-1988 of China adult physical dimension, adopt the human body standard of southern china people the 95th percentile, set up the restriction range of design variable; Employing dynamically reduces the method design of inertia weight based on the improvement particle swarm optimization of simulated annealing, and the optimal solution that finally obtains is as follows: S ₁=426.7, S ₂=342.5, S ₃=86.9.

Work process of the present invention

Can see line segment O in conjunction with Fig. 1 ₁O ₂₂Be ectoskeleton thigh model, O ₂O ₂₂Be ectoskeleton shank model, O ₂₁O ₂₃Be the ectoskeleton Hydraulic Cylinder Model, the last hinge that is connected between hydraulic cylinder and the thigh is O ₂₁, the following hinge that is connected between hydraulic cylinder and the shank is O ₂₃, the distance between last hinge and the knee joint is S ₁, the distance between following hinge and the knee joint is S ₃, the distance between the upper and lower hinge is S ₂, S ₁, S ₂And S ₃Be the parameter of finding the solution of ectoskeleton optimal design; With thigh upper extreme point O ₁The stressed power F that is reduced to a vertical direction, entire mechanism is carried out force analysis, can obtain F _CylinderFunctional relationship with F makes F _Cylinder=0 is optimization aim, adopts China adult size to set up the restriction range of variable; In order to strengthen the optimizing ability of particle swarm optimization algorithm, adopt the mode that dynamically reduces inertia weight ω that the standard particle swarm optimization is improved, by to the variable assignments in the algorithm, finally find the solution and obtain three location parameters between the hinge.

Claims (1)

1. an assistance type ectoskeleton knee joint tri-joint mechanism optimization method is optimized processing to assistance type ectoskeleton knee joint parameter, it is characterized in that, comprises following steps:

1) set up the position relation of joint ectoskeleton knee joint tri-joint mechanism: tri-joint mechanism comprises knee joint O ₂₂, the last hinge O that is connected with thigh of hydraulic cylinder ₂₁The following hinge O that is connected with shank with hydraulic cylinder ₂₃, the distance between last hinge and the knee joint is the long S of bar ₁, the distance between following hinge and the knee joint is the long S of bar ₃, the distance between the upper and lower hinge is the long S of bar ₂

2) by the improvement micropartical colony optimization algorithm based on simulated annealing, be optimization aim with the loaded hydraulic cylinder minimum, finally obtain the long S of bar ₁, the long S of bar ₂With the long S of bar ₃Value, and this value imported follow-up manufacturing process controlled.

Images