About

Max Shepherd is an Assistant Professor with a joint appointment in Mechanical and Industrial Engineering, and Physical Therapy and Rehabilitation at Northeastern University College of Engineering. His research focuses on designing and controlling wearable robotics, such as prosthetics and exoskeletons, used to rehabilitate and assist individuals with mobility impairments. His work spans gait biomechanics, machine learning, robotics, mechatronic design, and human motor control and perception. Shepherd's lab aims to develop personalized gait rehabilitation systems through innovative robotic devices. Prior to joining Northeastern, Dr. Shepherd completed his PhD in Biomedical Engineering at Northwestern University, conducting research at the Shirley Ryan AbilityLab, and was a Postdoctoral researcher at Georgia Tech. He has also worked at X (formerly Google X) and served as a visiting scholar at Ossur, an Icelandic prosthetics manufacturer. His contributions include developing adaptive prosthetic and exoskeleton technologies, such as variable-stiffness ankle prostheses and task-agnostic exoskeleton control systems, with the goal of improving mobility and comfort for users with mobility impairments.

Research topics

• Computer Science
• Artificial Intelligence
• Anatomy
• Mathematics
• Medicine
• Physics
• Simulation
• Physical medicine and rehabilitation

Selected publications

• Ankle Exoskeleton Control via Data-Driven Gait Estimation for Walking, Running, and Inclines

  IEEE Robotics and Automation Letters · 2025-04-16 · 2 citations

  articleSenior author

  Ankle exoskeletons have the potential to augment mobility, but control strategies have largely failed to seamlessly adapt to changes in the locomotion task. Here, we introduce a multi-headed network that predicts gait speed, ground incline, stance/swing transitions, and percent stance. These predictions are mapped to exoskeleton torque using typical biological torques as a guide. The model was trained on 9 subjects walking/jogging for 12 minutes across a range of speeds and inclines. The controller was validated on 4 subjects, and achieved stance phase prediction error of 3.4% across a range of speeds and inclines, both inside and outside the training set distribution. A secondary analysis showed similar accuracy could have been obtained with only 10% of the collected data, suggesting researchers may need fewer total strides of training data, provided the data is sufficiently diverse across users and tasks. Metabolic cost was improved during running compared to wearing the exoskeleton powered off, but was beneficial for only one subject during level walking and ramp ascent when compared to no exoskeleton. Overall, our controller smoothly adapted to time-varying inclines and walking/jogging speeds, and achieved high accuracy with a reduced training dataset, though larger torque magnitudes may be required to see metabolic benefit.

  PublisherDOI

• Uncertainty-Aware Ankle Exoskeleton Control

  ArXiv.org · 2025-08-28

  preprintOpen accessSenior author

  Lower limb exoskeletons show promise to assist human movement, but their utility is limited by controllers designed for discrete, predefined actions in controlled environments, restricting their real-world applicability. We present an uncertainty-aware control framework that enables ankle exoskeletons to operate safely across diverse scenarios by automatically disengaging when encountering unfamiliar movements. Our approach uses an uncertainty estimator to classify movements as similar (in-distribution) or different (out-of-distribution) relative to actions in the training set. We evaluated three architectures (model ensembles, autoencoders, and generative adversarial networks) on an offline dataset and tested the strongest performing architecture (ensemble of gait phase estimators) online. The online test demonstrated the ability of our uncertainty estimator to turn assistance on and off as the user transitioned between in-distribution and out-of-distribution tasks (F1: 89.2). This new framework provides a path for exoskeletons to safely and autonomously support human movement in unstructured, everyday environments.

  PublisherOA PDFDOI

• The Footropter: A Passive Prosthetic Prescription Tool With Adjustable Forefoot and Hindfoot Stiffness

  IEEE Transactions on Neural Systems and Rehabilitation Engineering · 2025-01-01 · 2 citations

  articleOpen accessSenior author

  Commercially available prosthetic feet are fabricated to have a fixed forefoot and hindfoot stiffness that cannot be changed in a clinical setting. This does not allow for patients to quickly compare multiple prosthetic foot stiffnesses to choose the stiffness they like the most while walking. In this paper, we present the Footropter, a passive prosthetic foot prescription tool that allows Certified Prosthetists (CPs) to rapidly change both the forefoot and hindfoot stiffnesses. The forefoot stiffness is changed by repositioning a spring clamp along a length of unbonded fiberglass layers and the hindfoot stiffness is changed by repositioning a single heel spring support. We introduce the design and working principles, characterize the ranges of available forefoot and hindfoot stiffnesses, and demonstrate the utility of the Footropter through two preference and perception studies with two unilateral transtibial prosthesis users. The Footropter, when paired with a preference optimization algorithm, can enable CPs to integrate patients' experiential input into the clinical prescription process.

  PublisherDOI

• Deep domain adaptation eliminates costly data required for task-agnostic wearable robotic control

  Science Robotics · 2025-11-19 · 1 citations

  article

  Data-driven methods have transformed our ability to assess and respond to human movement with wearable robots, promising real-world rehabilitation and augmentation benefits. However, the proliferation of data-driven methods, with the associated demand for increased personalization and performance, requires vast quantities of high-quality, device-specific data. Procuring these data is often intractable because of resource and personnel costs. We propose a framework that overcomes data scarcity by leveraging simulated sensors from biomechanical models to form a stepping-stone domain through which easily accessible data can be translated into data-limited domains. We developed and optimized a deep domain adaptation network that replaces costly, device-specific, labeled data with open-source datasets and unlabeled exoskeleton data. Using our network, we trained a hip and knee joint moment estimator with performance comparable to a best-case model trained with a complete, device-specific dataset [incurring only an 11 to 20%, 0.019 to 0.028 newton-meters per kilogram (Nm/kg) increase in error for a semisupervised model and 20 to 44%, 0.033 to 0.062 Nm/kg for an unsupervised model]. Our network significantly outperformed counterpart networks without domain adaptation (which incurred errors of 36 to 45% semisupervised and 50 to 60% unsupervised). Deploying our models in the real-time control loop of a hip/knee exoskeleton ( N = 8) demonstrated estimator performance similar to offline results while augmenting user performance based on those estimated moments (9.5 to 14.6% metabolic cost reductions compared with no exoskeleton). Our framework enables researchers to train real-time deployable deep learning, task-agnostic models with limited or no access to labeled, device-specific data.

  PublisherDOI

• Task-agnostic exoskeleton control via biological joint moment estimation

  Nature · 2024-11-13 · 74 citations

  article
  PublisherDOI

• The Variable Stiffness Orthosis: Customizable Mechanics for Assistance and Rehabilitation

  2024-11-27 · 2 citations

  preprintOpen access

  Challenges with community mobility are among the most prevalent disabilities worldwide, yet the current standard of care-passive orthotics-have remained largely unchanged for hundreds of years. Powered orthoses have been developed to address these shortcomings, but challenges with reliability, safety, weight, noise, and cost have stunted commercial translation. In this work, we present the Variable Stiffness Orthosis (VSO), a quasi-passive, ankle-foot orthosis that strikes a balance between powered and passive orthoses in terms of functionality and commercial practicality. The VSO can render customized, nonlinear torque-angle relationships via passive cam-based modules, which can feature extreme or even negative stiffness. A passive cam switching mechanism also decouples energy storage and return, allowing push off to be augmented with energy recycled from early stance phase, and changing equilibrium angle to simultaneously promote swing-phase foot clearance and standing stability. The VSO also features step-to-step stiffness adjustments spanning the softest to stiffest commercial AFOs via a motorized spring support. Pilot testing was performed on two participants with and without sciatic nerve injury (SNI). Both participants had activity-dependent stiffness preferences that spanned a large stiffness range. Preliminary results showed that using the VSO led to reduced foot drop, increased self-selected speed, increased total ankle moments, reduced biological moments, reduced toe striking, and reduced steppage in the participant with SNI compared to daily-use AFO and shoes-only conditions.

  PublisherDOI

• Improving Biological Joint Moment Estimation During Real-World Tasks With EMG and Instrumented Insoles

  IEEE Transactions on Biomedical Engineering · 2024-04-15 · 10 citations

  articleOpen access

  OBJECTIVE: Real-time measurement of biological joint moment could enhance clinical assessments and generalize exoskeleton control. Accessing joint moments outside clinical and laboratory settings requires harnessing non-invasive wearable sensor data for indirect estimation. Previous approaches have been primarily validated during cyclic tasks, such as walking, but these methods are likely limited when translating to non-cyclic tasks where the mapping from kinematics to moments is not unique. METHODS: We trained deep learning models to estimate hip and knee joint moments from kinematic sensors, electromyography (EMG), and simulated pressure insoles from a dataset including 10 cyclic and 18 non-cyclic activities. We assessed estimation error on combinations of sensor modalities during both activity types. RESULTS: Compared to the kinematics-only baseline, adding EMG reduced RMSE by 16.9% at the hip and 30.4% at the knee (p < 0.05) and adding insoles reduced RMSE by 21.7% at the hip and 33.9% at the knee (p < 0.05). Adding both modalities reduced RMSE by 32.5% at the hip and 41.2% at the knee (p < 0.05) which was significantly higher than either modality individually (p < 0.05). All sensor additions improved model performance on non-cyclic tasks more than cyclic tasks (p < 0.05). CONCLUSION: These results demonstrate that adding kinetic sensor information through EMG or insoles improves joint moment estimation both individually and jointly. These additional modalities are most important during non-cyclic tasks, tasks that reflect the variable and sporadic nature of the real-world. SIGNIFICANCE: Improved joint moment estimation and task generalization is pivotal to developing wearable robotic systems capable of enhancing mobility in everyday life.

  PublisherOA PDFDOI

• Mitigating Crouch Gait With an Autonomous Pediatric Knee Exoskeleton in the Neurologically Impaired

  Journal of Biomechanical Engineering · 2024-08-28 · 3 citations

  article

  Crouch gait is one of the most common compensatory walking patterns found in individuals with neurological disorders, often accompanied by their limited physical capacity. Notable kinematic characteristics of crouch gait are excessive knee flexion during stance and reduced range of motion during swing. Knee exoskeletons have the potential to improve crouch gait by providing precisely controlled torque assistance directly to the knee joint. In this study, we implemented a finite-state machine-based impedance controller for a powered knee exoskeleton to provide assistance during both stance and swing phases for five children and young adults who exhibit chronic crouch gait. The assistance provided a strong orthotic effect, increasing stance phase knee extension by an average of 12 deg. Additionally, the knee range of motion during swing was increased by an average of 15 deg. Changes to spatiotemporal outcomes, such as preferred walking speed and percent stance phase, were inconsistent across subjects and indicative of the underlying intricacies of user response to assistance. This study demonstrates the potential of knee exoskeletons operating in impedance control to mitigate the negative kinematic characteristics of crouch gait during both stance and swing phases of gait.

  PublisherDOI

• Rethinking Energy Storage and Return in Prosthetic Feet: User Preferences Challenge Conventional Wisdom

  2024-09-01

  article1st authorCorresponding

  Modern prosthetic feet have spring-like mechanics, deflecting and storing energy during mid-stance, and returning this energy during terminal stance. Researchers and manufacturers of prosthetic feet often tout the high energy storage and return of new prosthetic designs, but there is limited evidence supporting the notion that more energy storage and return is best for the user in terms of either biomechanical outcomes or user preference. In this paper, the relationship between ankle stiffness and energy storage is evaluated at stiffness levels within ± 20% of the user-preferred (self-selected) stiffness. For all eight amputee subjects, energy storage and return was highest at stiffness levels below the preferred stiffness. These results indicate that maximal energy storage and return may occur at an uncomfortably low stiffness, casting doubt on the utility of energy storage and return as a metric for evaluating prosthetic designs.

  PublisherDOI

• Faster-than-reflexes robo-boots boost balance

  2023-02-15

  preprintSenior author
  PublisherDOI

Frequent coauthors

• Elliott J. Rouse

  University of Michigan–Ann Arbor

  36 shared

• Tyler R. Clites

  University of California, Los Angeles

  15 shared

• Levi J. Hargrove

  Northwestern University

  8 shared

• Ann M. Simon

  Shirley Ryan AbilityLab

  8 shared

• Dean D. Molinaro

  Georgia Institute of Technology

  7 shared

• Aaron J. Young

  Georgia Institute of Technology

  7 shared

• Daniel Gunz

  MED-EL (Austria)

  6 shared

• Christophe Lecomte

  6 shared

Labs

• Shepherd Lab: Robotics - Rehabilitation - LocomotionPI

Awards & honors

• Fall 2024 Spark Fund Award from Northeastern’s Center for Re…