Neuromechanical Model-Based Control for Soft Exosuit Back Assistance
The paper "Adaptive assistance with an active and soft back-support exosuit to unknown external loads via model-based estimates of internal lumbosacral moments" by Alejandro Moya-Esteban et al. presents a novel approach to controlling back-support exosuits through a neuromechanical model-based controller (NMBC). Rather than relying on traditional kinematic measurements, this exosuit leverages real-time electromyography (EMG)-driven models to inform force assistance, addressing a gap in adaptive support during lifting tasks with unknown external loads.
Overview and Key Contributions
The paper introduces a NMBC for a soft exosuit designed to provide adaptive support by modulating assistive forces based on the active component of lumbosacral joint moments. These moments are estimated using real-time EMG data, which inherently adapts to variations in external loads such as differing weights of lifted objects. This methodology is set apart from conventional virtual spring-based control (VSBC) strategies, which depend on kinematic indicators such as trunk inclination and do not adapt to variations in load conditions.
The experimental paper with ten participants demonstrated that the NMBC could better modulate assistive forces than the VSBC when participants lifted 5 kg and 15 kg weights. Notably, the NMBC achieved superior reductions in cumulative compression forces at the lumbosacral joint, with improvements of 18.2% and 21.3% for the lower and higher weight conditions, respectively, compared to 10.7% and 10.2% with VSBC.
Numerical Results and Implications
Importantly, the NMBC modulated peak assistive forces to 2.13 N/kg for the 5 kg and 2.82 N/kg for the 15 kg conditions, while VSBC forces remained relatively constant, evidencing the NMBC's adaptive capability. This dynamic adjustment potentially allows the exosuit to support users more effectively across varying biomechanical demands, translating to a broader scope of practical application in occupational settings.
The results indicate that the NMBC could provide meaningful reductions in muscular effort and spinal loading, important factors in occupational health. Given that low back pain (LBP) affects a significant portion of the global workforce, the development of adaptive exoskeleton assistance technologies holds promise for reducing the prevalence of work-related musculoskeletal disorders.
Theoretical and Practical Implications
From a theoretical perspective, this research underscores the potential of EMG-driven models in configuring human-machine interfaces that adapt to user-specific and task-specific conditions. Practically, the implementation of the NMBC in soft exosuits could enhance their usability in various domains, from industrial settings to healthcare environments. The ability to provide load-adaptive and subject-specific support could lead to greater acceptance and longer adoption spans for such wearable robotics.
Future Directions
Future research could explore the integration of NMBCs in various exosuit configurations, extending beyond the lumbar support scope to other applications such as upper-limb assistance. Furthermore, refinement of the EMG signal processing and real-time model execution could enhance responsiveness and increase user comfort. Investigating real-world deployments would also be invaluable for examining long-term benefits and ergonomics.
Overall, the paper presents a sophisticated evolutionary step in wearable robotics for occupational health, emphasizing the crucial interplay between biological signals and mechanical assistance in optimizing human-robot interactions.