Robust Torque Predictions From Electromyography Across Multiple Levels of Active Exoskeleton Assistance Despite Non-linear Reorganization of Locomotor Output

Robotic exoskeletons can assist humans with walking by providing supplemental torque in proportion to the user's joint torque. Electromyographic (EMG) control algorithms can estimate a user's joint torque directly using real-time EMG recordings from the muscles that generate the torque. However, EMG signals change as a result of supplemental torque from an exoskeleton, resulting in unreliable estimates of the user's joint torque during active exoskeleton assistance. Here, we present an EMG control framework for robotic exoskeletons that provides consistent joint torque predictions across varying levels of assistance. Experiments with three healthy human participants showed that using diverse training data (from different levels of assistance) enables robust torque predictions, and that a convolutional neural network (CNN), but not a Kalman filter (KF), can capture the non-linear transformations in EMG due to exoskeleton assistance. With diverse training, the CNN could reliably predict joint torque from EMG during zero, low, medium, and high levels of exoskeleton assistance [root mean squared error (RMSE) below 0.096 N-m/kg]. In contrast, without diverse training, RMSE of the CNN ranged from 0.106 to 0.144 N-m/kg. RMSE of the KF ranged from 0.137 to 0.182 N-m/kg without diverse training, and did not improve with diverse training. When participant time is limited, training data should emphasize the highest levels of assistance first and utilize at least 35 full gait cycles for the CNN. The results presented here constitute an important step toward adaptive and robust human augmentation via robotic exoskeletons. This work also highlights the non-linear reorganization of locomotor output when using assistive exoskeletons; significant reductions in EMG activity were observed for the soleus and gastrocnemius, and a significant increase in EMG activity was observed for the erector spinae. Control algorithms that can accommodate spatiotemporal changes in muscle activity have broad implications for exoskeleton-based assistance and rehabilitation following neuromuscular injury.

[1]  Tyler Davis,et al.  Portable Take-Home System Enables Proportional Control and High-Resolution Data Logging With a Multi-Degree-of-Freedom Bionic Arm , 2020, Frontiers in Robotics and AI.

[2]  Daniel P Ferris,et al.  Locomotor adaptation to a soleus EMG-controlled antagonistic exoskeleton. , 2013, Journal of neurophysiology.

[3]  Zhongcai Pei,et al.  Development of an EMG-Controlled Knee Exoskeleton to Assist Home Rehabilitation in a Game Context , 2019, Front. Neurorobot..

[4]  Marshall K. Ishmael,et al.  ExoProsthetics: Assisting Above-Knee Amputees with a Lightweight Powered Hip Exoskeleton , 2019, 2019 IEEE 16th International Conference on Rehabilitation Robotics (ICORR).

[5]  Gregory A. Clark,et al.  Inexpensive and Portable System for Dexterous High-Density Myoelectric Control of Multiarticulate Prostheses , 2020, 2020 IEEE International Conference on Systems, Man, and Cybernetics (SMC).

[6]  Rong Song,et al.  Movement Performance of Human–Robot Cooperation Control Based on EMG-Driven Hill-Type and Proportional Models for an Ankle Power-Assist Exoskeleton Robot , 2017, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[7]  F C T van der Helm,et al.  Inverse dynamics calculations during gait with restricted ground reaction force information from pressure insoles. , 2006, Gait & posture.

[8]  Monica A. Daley,et al.  A Physiologist's Perspective on Robotic Exoskeletons for Human Locomotion , 2007, Int. J. Humanoid Robotics.

[9]  Daniel P. Ferris,et al.  Robotic lower limb exoskeletons using proportional myoelectric control , 2009, 2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[10]  Sunil K. Agrawal,et al.  Powered Hip Exoskeletons Can Reduce the User's Hip and Ankle Muscle Activations During Walking , 2013, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[11]  V. John Mathews,et al.  Polynomial Kalman filter for myoelectric prosthetics using efficient kernel ridge regression , 2017, 2017 8th International IEEE/EMBS Conference on Neural Engineering (NER).

[12]  Jonathan Camargo,et al.  Biological Hip Torque Estimation using a Robotic Hip Exoskeleton , 2020, 2020 8th IEEE RAS/EMBS International Conference for Biomedical Robotics and Biomechatronics (BioRob).

[13]  Scott Kuindersma,et al.  Human-in-the-loop optimization of hip assistance with a soft exosuit during walking , 2018, Science Robotics.

[14]  Gregory A. Clark,et al.  Improved Training Paradigms and Motor-decode Algorithms: Results from Intact Individuals and a Recent Transradial Amputee with Prior Complex Regional Pain Syndrome , 2018, 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[15]  C. Walsh,et al.  A soft robotic exosuit improves walking in patients after stroke , 2017, Science Translational Medicine.

[16]  Conor J. Walsh,et al.  Walking Faster and Farther With a Soft Robotic Exosuit: Implications for Post-Stroke Gait Assistance and Rehabilitation , 2020, IEEE Open Journal of Engineering in Medicine and Biology.

[17]  V. Miller A Physiologist’s Perspective , 2010 .

[18]  Jun Morimoto,et al.  EMG-Based Model Predictive Control for Physical Human–Robot Interaction: Application for Assist-As-Needed Control , 2018, IEEE Robotics and Automation Letters.

[19]  Gregory A. Clark,et al.  Intuitive neuromyoelectric control of a dexterous bionic arm using a modified Kalman filter , 2019, Journal of Neuroscience Methods.

[20]  M. Ferrarin,et al.  A multiple-task gait analysis approach: kinematic, kinetic and EMG reference data for healthy young and adult subjects. , 2011, Gait & posture.

[21]  Ignacio Galiana,et al.  Reducing the metabolic rate of walking and running with a versatile, portable exosuit , 2019, Science.

[22]  Nicola Vitiello,et al.  Intention-Based EMG Control for Powered Exoskeletons , 2012, IEEE Transactions on Biomedical Engineering.

[23]  Wei Wu,et al.  Bayesian Population Decoding of Motor Cortical Activity Using a Kalman Filter , 2006, Neural Computation.

[24]  Michael E. Hahn,et al.  A NEURAL NETWORK MODEL FOR ESTIMATION OF NET JOINT MOMENTS DURING NORMAL GAIT , 2008 .

[25]  Tyler Davis,et al.  Portable Take-Home System Enables Proportional Control and High-Resolution Data Logging With a Multi-Degree-of-Freedom Bionic Arm , 2020, bioRxiv.

[26]  A. d’Avella,et al.  EMG patterns during assisted walking in the exoskeleton , 2014, Front. Hum. Neurosci..

[27]  Luciano Luporini Menegaldo,et al.  Real-time muscle state estimation from EMG signals during isometric contractions using Kalman filters , 2017, Biological Cybernetics.

[28]  Activities of daily living with bionic arm improved by combination training and latching filter in prosthesis control comparison , 2021, Journal of neuroengineering and rehabilitation.