Skip to main content

Spinal Exoskeletal Robot for Low Back Pain Prevention and Vocational Reintegration

Periodic Reporting for period 3 - SPEXOR (Spinal Exoskeletal Robot for Low Back Pain Prevention and Vocational Reintegration)

Reporting period: 2019-01-01 to 2019-12-31

Low-back pain is the leading cause of worker absenteeism after the common cold, accounting for 15% of sick leaves and hundreds of millions of lost work days annually. Most of today’s robotic assistive devices are in forms of exoskeletons that augment the motion of legs and arms and neglect the role of spinal column in transferring load from the upper body and arms to the legs. In SPEXOR we will filled this gap and designed a novel and revolutionary spinal exoskeleton to prevent low-back pain in able bodied workers and to support workers with low-back pain in vocational rehabilitation.
The concept of SPEXOR is driven by several interdisciplinary ideas that push current understanding of low-back pain intervention through several innovative research and technological stages. First, robot-centred requirements for low-back pain prevention are determined and a musculoskeletal stress monitoring system is developed to unobtrusively measure the associated key variables. Then, optimal design parameters and sensorimotor strategies are provided with respect to the robot-centred requirements and their associated key variables. Based on these aspects, a spinal exoskeleton mechanism and its actuation is developed and adaptive control architecture is employed. Such research and development cycle is enclosed by multi-phase end-user evaluation, usability and satisfaction studies.
The project builds upon the partner’s extensive experience with work ergonomics, modelling and optimization of human movement, design, control and evaluation of exoskeletons. Several beyond-the-state-of-art scientific approaches and technologies were employed through a colorful mixture of research, industrial, SME and end-user partners of the consortium.
Ultimately, the results of SPEXOR will have a significant impact well beyond the current scientific understanding and technological capabilities of assistive robots used in daily life and health care.
Benchmark testing was performed using market available devices, and biomechanical requirements were formulated. The SPEXOR passive device was evaluated using participants familiar with professional load handling. The sensory system was evaluated using laboratory equipment. Limits to accuracy in moment estimates are mainly due to the technical performance of force insoles.
A musculoskeletal stress monitoring system was implemented and optimized using a rigid body model to compute the joint kinematics and torques of the trunk using wireless inertial sensors and foot soles. Real time feedback on key spine load variables was developed, and several feedback source signals were tested.
We developed a series of human models and parameterized exoskeleton models and solved optimal control problem formulations to contribute to a better understanding of the complex dynamic interactions of the human body and different version of the exoskeletons supporting the current developments in the project. Prediction models for the behavior of persons with and without back pain have been improved by means of inverse optimal control based on appropriate experimental studies.
We developed two quasi-passive back support exoskeletons with special combination of joints, that compensate the misalignment between the joints of the exoskeleton and the joints of the user, while at the same time the exoskeleton unloads the lower back by redirecting the unwanted forces and moments from the lumbo-sacral spinal region to the less harmful and more stable location on the pelvis, thighs and chest. Besides, we developed two active exoskeletons with a redesigned spinal module, a novel electrical and hydraulic actuation.
A novel control concept was developed, implemented and evaluated for the quasi-passive spinal exoskeletons and for both active exoskeletons. The control concept is based on probabilistic state classifiers combined with a state machine and transition phases.
End user requirements and expectations regarding benefits of a spinal exoskeleton for people in load handling professions and low back pain rehabilitation were obtained. Twenty-four employees, with and without a history of low back pain, from two big international companies were recruited for testing the first prototype of the passive SPEXOR spinal exoskeleton.
It was found that the functionality and support to the spine for the SPEXOR system was better than for currently available systems. This suggests an impact on the risk of low back pain due to spine overloading in workers.
The novelty of the musculoskeletal stress monitoring system lies in its real time bidirectional interface that allows real-time transfer of wireless sensor data to calculate kinematic and dynamic values of the user’s posture and give feedback of the most relevant information back to the user. Preliminary testing of feedback suggests that participants modify their lifting behavior specifically in response to these signals.
The simulation approaches developed in SPEXOR significantly contribute to a better understanding of critical motions and loads the result in an increase back pain and to the design of exoskeletons with motion intelligence that will result in an increased support of the user as well as a higher comfort while requiring a reduced effort in terms of prototype iterations.
Comparing to the rigid devices, the SPEXOR device offers 20 % larger range of motion allowing users to reach the floor with the finger tips while wearing the exoskeleton. Additionally, wearability and easiness-of-use, which are the two important technological barriers that often reduce acceptability of exoskeletons among the healthy individuals, were significantly improved.
We developed a novel control algorithm brings contributions to the problem of estimating and predicting the human motion activates through the sensory feedback. Consequently, the proposed method improves the timing and reliability of motion classification in real-time. Hence, the overall performance of the passive spinal exoskeleton has been improved.
The SPEXOR passive exoskeleton was demonstrated to significantly reduce the mechanical load and muscle activity at the low back during lifting and trunk bending, and reduced aerobic load during repetitive lifting. It improved objective and subjective performance and low back discomfort during lifting and bending with minor hindrance of other functional task in a mock up test battery. Moreover, it was shown that people with a history of low back pain reported a higher self-efficacy for performing daily activities after having experienced the potential support of the SPEXOR exoskeleton. The observed effects on biomechanical support, functional performance and user-satisfaction all exceed previous benchmark tests with similar test protocols of a current commercially available device. The development was rated positive by the companies from which end-users were recruited and additional test at these work sites are currently scheduled.
Different exoskeleton prototype models have been optimized in the SPEXOR project
SPEXOR monitoring system used for a lifting scenario with real-time biomechanical model
Flexible spine, missalignment mechanism and passive hip module
Participant bending forward while wearing the SPEXOR passive system during testing
Functional performance testing of the SPEXOR passive exoskeleton at a Dutch automotive company
Test battery sample demonstrating exoskeleton provided torque when forward bending