Recent advancements in bionic prosthetic technology offer transformative opportunities to restore mobility and functionality for individuals with missing limbs. Users of bionic limbs, or bionic humans, learn to seamlessly integrate prosthetic extensions into their motor repertoire, regaining critical motor abilities. The remarkable movement generalization and environmental adaptability demonstrated by these individuals highlight motor intelligence capabilities unmatched by current artificial intelligence systems. Addressing these limitations, MyoChallenge '24 at NeurIPS 2024 established a benchmark for human-robot coordination with an emphasis on joint control of both biological and mechanical limbs. The competition featured two distinct tracks: a manipulation task utilizing the myoMPL model, integrating a virtual biological arm and the Modular Prosthetic Limb (MPL) for a passover task; and a locomotion task using the novel myoOSL model, combining a bilateral virtual biological leg with a trans-femoral amputation and the Open Source Leg (OSL) to navigate varied terrains. Marking the third iteration of the MyoChallenge, the event attracted over 50 teams with more than 290 submissions all around the globe, with diverse participants ranging from independent researchers to high school students. The competition facilitated the development of several state-of-the-art control algorithms for bionic musculoskeletal systems, leveraging techniques such as imitation learning, muscle synergy, and model-based reinforcement learning that significantly surpassed our proposed baseline performance by a factor of 10. By providing the open-source simulation framework of MyoSuite, standardized tasks, and physiologically realistic models, MyoChallenge serves as a reproducible testbed and benchmark for bridging ML and biomechanics. The competition website is featured here: https://sites.google.com/view/myosuite/myochallenge/myochallenge-2024.

Owing to the restrictions of live experimentation, musculoskeletal simulation models play a key role in biological motor control studies and investigations. Successful results of which are then tried on live subjects to develop treatments as well as robot aided rehabilitation procedures for addressing neuromusculoskeletal anomalies ranging from limb loss, to tendinitis, from sarcopenia to brain and spinal injuries. Despite its significance, current musculoskeletal models are computationally expensive, and provide limited support for contact-rich interactions which are essential for studying motor behaviors in activities of daily living, during rehabilitation treatments, or in assistive robotic devices. To bridge this gap, this work proposes an automatic pipeline to generate physiologically accurate musculoskeletal, as well as hybrid musculoskeletal-exoskeletal models. Leveraging this pipeline we present MyoSim - a set of computationally efficient (over 2 orders of magnitude faster than state of the art) musculoskeletal models that support fully interactive contact rich simulation. We further extend MyoSim to support additional features that help simulate various real-life changes/diseases, such as muscle fatigue, and sarcopenia. To demonstrate the potential applications, several use cases, including interactive rehabilitation movements, tendon-reaffirmation, and the cosimulation with an exoskeleton, were developed and investigated for physiological correctness. Web-page: https://sites.google.com/view/myosuite

To enable the broad adoption of wearable robotic exoskeletons in medical and industrial settings, it is crucial they can adaptively support large repertoires of movements. We propose a new human-machine interface to simultaneously drive bilateral ankle exoskeletons during a range of “unseen” walking conditions and transitions that were not used for establishing the control interface. The proposed approach used person-specific neuromechanical models to estimate biological ankle joint torques in real-time from measured electromyograms (EMGS) and joint angles. We call this “neuromechanical model-based control” (NMBC). NMBC enabled six individuals to voluntarily control a bilateral ankle exoskeleton across six walking conditions, including all intermediate transitions, i.e., two walking speeds, each performed at three ground elevations. A single subject case-study was carried out on a dexterous locomotion tasks involving moonwalking. NMBC always enabled reducing biological ankle torques, as well as eight ankle muscle EMGs both within (22% torque;12% EMG) and between walking conditions (24% torque; 14% EMG) when compared to non-assisted conditions. Torque and EMG reductions in novel walking conditions indicated that the exoskeleton operated symbiotically, as an exomuscle controlled by the operator.s neuromuscular system. This opens new avenues for the systematic adoption of wearable robots as part of out-of-the-lab medical and occupational settings.

With this work we introduce a novel methodology for the simulation of walking of a humanoid robot. Motion capture technology is used to calibrate the dynamics engine internal parameters and validate the simulated motor task. Results showed the calibrated contact model allows predicting dynamically consistent ground reaction forces and producing realistic whole-body kinematics.

The inclusion of robotic systems in physiotherapy allows developing new solutions for the rehabilitation and support of disabled people. Our research addresses the core problem for the advancement of such applications: the availability of a human machine interface offering intuitive control of robotic devices. In this paper we present an EMG-driven model of the human lower limb based on that previously developed by Lloyd et al. We then introduce a set of enhancements that allow reducing time and memory requirements and provide real-time performances for the control of a lower limb powered orthosis.

The next generation of tools for rehabilitation robotics requires advanced human-robot interfaces able to activate the device as soon as patient's motion intention is raised. This paper investigated the suitability of Support Vector Machine (SVM) classifiers for identification of locomotion intentions from surface electromyography (sEMG) data. A phase-dependent approach, based on foot contact and foot push off events, was employed in order to contextualize muscle activation signals. Good accuracy is demonstrated on experimental data from three healthy subjects. Classification has also been tested for different subsets of EMG features and muscles, aiming to identify a minimal setup required for the control of an EMG-based exoskeleton for rehabilitation purposes.