Reducing the muscle activity of walking using a portable hip exoskeleton based on human-in-the-loop optimization

Front. Bioeng. Biotechnol., 2023 · DOI: 10.3389/fbioe.2023.1006326 · Published: May 4, 2023

Assistive Technology Biomedical Biomechanics

Simple Explanation

This paper explores how to make wearable robots, specifically hip exoskeletons, work better by using a method called human-in-the-loop optimization. This method involves adjusting the way the exoskeleton helps based on real-time feedback from the person wearing it, specifically by monitoring muscle activity. The goal is to reduce the amount of effort the person needs to exert while walking, making the exoskeleton more efficient and comfortable.

Study Duration

Not specified

Participants

Four volunteers

Evidence Level

Not specified

Key Findings

1
Human-in-the-loop optimization led to muscle activity reduction of 33.56% and 41.81% at most when compared to walking without and with the hip exoskeleton, respectively.
2
Three out of four participants achieved superior outcomes compared to the predeﬁned assistance patterns using human-in-the-loop optimization.
3
The order of the two typical optimizers, i.e., Bayesian and CMA-ES, did not affect the optimization results.

Research Summary

This study introduces a muscle-activity-based human-in-the-loop (HIL) optimization strategy for a portable hip exoskeleton to reduce the time spent on collecting biosignals during each iteration. The effectiveness of both Bayesian Optimization (BO) and Covariance Matrix Adaptive Evolution Strategy (CMA-ES) algorithms was validated and compared through experimental results with four volunteers. The results demonstrated that the HIL optimization strategy based on sEMG signals can lower the time investment of each iteration and generate superior assistance patterns compared to predeﬁned ones.

Practical Implications

Improved Exoskeleton Design

The study provides insights into designing more effective and personalized hip exoskeletons for assisting walking, potentially benefiting individuals with mobility impairments.

Faster Optimization Strategies

The muscle-activity-based HIL optimization can reduce the time required for customizing assistive torque patterns, making it more practical for real-world applications.

Wider Application

The findings can guide the optimization of assistive and resistive torque/force profiles for upper/lower limb wearable devices, benefiting augmentation and physical training.

Study Limitations

1
Only the RF muscle activity was measured for HIL optimization of torque patterns.
2
Only a single walking scenario, i.e., treadmill walking with constant speed, was tested.
3
Only four subjects were recruited for the trials.

Your Feedback

Was this summary helpful?

Back to Assistive Technology