Characteristic Analysis And Modeling On The Muscular Force Of The Lower Limb During Human Standing Balance

Muscle is the source of the human movement, and it is one of the main contents ofbiomechanics and bionics to research the dynamic characteristics of muscle force duringhuman movement. The further study on the working mechanism and the coordinationmechanism of muscle,which can contribute to expanding the biomechanical researchcategory and promoting the development of engineering, bio-medical, sports and otherdisciplines. As an important carrier of the multi-joint and multi-muscle, the human bodyhas been researched by many experts for a long time. However, due to the limitations ofhuman activity, diversity and measuring instruments, it is more difficult to calculate themuscle force directly by means of the experimental methods. In this paper, combining thedynamic model with the optimization theory, the distributions of the main muscle force inthe lower limb are predicted indirectly during human standing balance under the passivemovement, which provide strong theoretical guidance and help for the development ofrehabilitation medicine and intelligence prosthetic.Firstly, in view of the adjustment mechanism of human standing balance, themusculoskeletal system of lower limb is simplified to a physical model with3joints and9muscles. Then on the basis of this model, an optimum mathematical model is built up tosolve the problem of redundant muscle forces. Moreover, classical lagrangian modelingmethod is adopted to build a dynamic model of human lower limb, and gain the jointtorques by dynamic analysis.Then, starting from the optimization method, the particle swarm optimization withthe rapid convergence is chosen to estimate the redundant muscle forces of humanstanding balance. In consideration of the limit that exists in the process with manyobjective functions and constraints, a multi-objective particle swarm optimizationalgorithm based on the hybrid penalty function(HMOPSO) is put forward, which has goodspeed and accuracy when dealing with the multi-objective optimization problem.Finally, aim at the three common objective functions of the muscle force prediction,the advantages, disadvantages and application condition of them are analyzed via simulation and experiments, then appropriate objective functions applied to the standingbalance are chosen. Furthermore, the proposed HMOPSO algorithm is used to calculatethe distribution and variation of the multi-objective muscle force. Simulation resultsobtained with the multi-objective particle swarm optimization algorithm show that it ismore aligned with the principle and coordination mechanism of the lower limb musclesduring human standing balance.