Session: 06-10-01: Robotics, Rehabilitation

Paper Number: 144358

144358 - Simulation Based Analysis and Design of Passive Back Exoskeleton for Squat Lifting 

Manual material handling tasks involve lifting heavy objects which results in several musculoskeletal disorders. Back pain is the most common disorder that occurs due to the lifting of loads. Two common techniques of lifting are squat and stoop lift. A lot of research is currently being focused on exoskeletons which have shown promising results in providing important assistance during lifting.  Passive exoskeletons such as Laevo, BackX (SuitX), active exoskeletons such as Xo trunk, and other back support devices have been developed to prevent back injuries by reducing the compressive forces and torques on the wearer’s lower back region (L5/S1 disc) while stooping, lifting objects and bending. However, a detailed analysis of the effect of exoskeleton on muscle activations and muscle forces is still needed. This work aims to investigate how the passive back exoskeleton affects the spinal muscle forces and spinal loads, through musculoskeletal simulation. Finding from such simulation can provide insights to further improve the back exoskeleton design.  For this purpose, the passive exoskeleton is modeled as a set of elastic elements (springs), and dynamic analysis for a lifting cycle is done during a squat lifting task.

 Musculoskeletal simulations were carried out in OpenSim software utilizing the open-source OpenSim lifting full body (LFB) model. This model comprises 30 segments, incorporates 238 Hill-type musculotendon actuators, and 29 degrees of freedom, trunk comprised eight rigid body segments (welded pelvis-sacrum, L5, L4, L3, L2, L1, and torso and six spherical joints located at the six intervertebral joints (T12/L1, L1/L2, L2/L3, L3/L4, L4/L5, L5/S1) and trunk musculature was modeled by eight major muscle groups: the erector spinae (ES), rectus abdominis (RA), external obliques (EO), internal obliques (IO), multifidus (MF), quadratus lumborum (QL), psoas major (PS), and latissimus dorsi (LD). The model has both upper and lower limb segments enhancing its ability to accurately represent body kinematics during lifting tasks. The LFB model is modified by adding an external spring that provides assistive forces as in a passive exoskeleton. Using open-source experimental data (marker data and external force data) of squat lifting task, inverse kinematics (IK) simulations are performed using this modified OpenSim model as input to obtain joint angles. Static Optimization simulations are then carried out using the IK results and external force data as input to determine muscular and spring forces during the lifting cycle. Joint reaction analysis is conducted to calculate forces and moments at joints, with a focus on the L5S1 joint.

 During the lifting cycle, erector spinea muscle and rectus abdominal play an important role in flexion and extension. During trunk extension, the erector spinea muscle acts as the agonist (prime mover) and provides extensor moment to the spine. During flexion (bending forward), the erector spinea muscle provides a moment to maintain postural stability. During the whole lifting cycle, the erector spine muscle is strained, so to unloading the erector spinea muscle, is the primary goal of most back support exoskeleton. The erector spinea muscles include the lumbar fibers (left) and corresponding muscle fascicles (right) for Longissimus thoracis pars lumborum (LTpl), Longissimus thoracis pars thoracis (LTpt), Iliocostalis lumborum pars lumborum (ILpL), Iliocostalis lumborum pars thoracis (ILpT).

Preliminary analysis is done by adding a path spring along the LTPL_L1_l (Longissimus thoracis pars lumborum _left) erector spinea muscle with stiffness 10kN/m and resting length 0.1m (as the length of LTPL_L1 muscle is 0.1m) results show that it decreases the muscle activation and muscle forces of erector spinea muscle during the lifting cycle. The maximum muscle force of LTPL_L1 decreases by 30 N . By adding a spring, the actuator moment also decreases during the lifting cycle. Overall aim of this work is to identify optimal spring configurations for design of passive back exoskeleton, to aid in squat lifting. Corresponding spring stiffnesses will also be determined. The optimized back exoskeleton is expected to reduce compressive and shear forces on the L5S1 joint during lifting.

 

Presenting Author: Neha Sahu Indian Institute of Technology Bombay

Presenting Author Biography: I am a 3rd year Ph.D. candidate in Mechanical Engineering at the Indian Institute of Technology, Bombay under Prof. Abhishek Gupta. I have done my MTech in Mechanical Engineering from IIT Dhanbad. I have done my B.tech Mechanical Engineering from B.I.E.T Jhansi. I am currently working on a project where I am developing the spine exoskeleton for assisting the spine during lifting tasks. I will perform a predictive study through OpenSim to analyze the effect of different design parameters and determine a control strategy. I have developed the code in OpenSim API for the spring attachment in the model.

Authors:

Neha Sahu Indian Institute of Technology Bombay
Abhishek Gupta Indian Institute of Technology Bombay