AI-Driven Hip Exoskeleton Control for Diverse Users and Walking Tasks

November 10, 2026 updated by: Inseung Kang, Carnegie Mellon University

AI-Driven Hip Exoskeleton Control Framework That Rapidly Generalizes to a Broad Range of Users and Real-World Locomotor Tasks

The goal of this clinical trial is to evaluate the feasibility of a prototype artificial intelligence (AI)-based controller for a lightweight hip exoskeleton across a broad range of users, including healthy younger adults (18-39 years) and older adults (65 years and older). The main questions it aims to answer are:

  • Can the AI-based controller accurately estimate gait phase in real time across younger and older adults?
  • How does hip exoskeleton assistance change walking energy cost, mechanics, and muscle activity?

Researchers will compare walking with powered hip exoskeleton assistance to walking without the exoskeleton and walking while wearing the exoskeleton with no assistance (zero torque).

Participants will:

  • Complete a short screening and frailty assessment
  • Walk on a treadmill under different conditions with and without the hip exoskeleton during 2 to 5 lab visits
  • Wear motion capture markers, muscle activity (EMG) sensors, and a mask that measures breathing to estimate energy use

Study Overview

Status

Recruiting

Conditions

Intervention / Treatment

Detailed Description

This study is funded by the National Institute of Biomedical Imaging and Bioengineering (R21EB037268) and conducted at the University of Texas at Austin in collaboration with Carnegie Mellon University. The CMU Hip Exoskeleton is a research prototype with powered hip joints for flexion and extension (peak torque 18 Nm, total weight 3.5 kg). Its controller uses joint angle and user-state estimates to command assistive torque, limited to less than 0.2 Nm/kg (at most about 30% of typical biological hip moments during walking), so it supplements rather than overrides the user's own muscle effort.

During each visit, participants walk on a force-instrumented treadmill at 1.25 m/s with safety harness and handrails. Researchers record whole-body motion capture, ground reaction forces, inertial measurement unit data, lower-limb electromyography, and metabolic energy expenditure via open-circuit indirect calorimetry. Walking conditions include no exoskeleton, exoskeleton worn with zero torque, and powered exoskeleton assistance. Outcomes will be compared across conditions and between younger and older adults using repeated-measures statistical analysis.

Study Type

Interventional

Enrollment (Estimated)

16

Phase

  • Not Applicable

Contacts and Locations

This section provides the contact details for those conducting the study, and information on where this study is being conducted.

Study Contact

Study Locations

Participation Criteria

Researchers look for people who fit a certain description, called eligibility criteria. Some examples of these criteria are a person's general health condition or prior treatments.

Eligibility Criteria

Ages Eligible for Study

18 years and older (Adult, Older Adult)

Accepts Healthy Volunteers

Yes

Genders Eligible for Study

All

Description

Inclusion Criteria:

  • Age 18 to 39 years (younger adult group) or 65 years and older (older adult group)
  • Willing and able to provide informed consent
  • Able to walk independently without assistance
  • No known cardiovascular, metabolic, or renal disease, and no signs or symptoms suggestive of such disease
  • Cleared for light- to moderate-intensity exercise without the need for medical clearance under American College of Sports Medicine (ACSM) preparticipation screening guidelines

Exclusion Criteria:

  • Age 40 to 64 years
  • Leg surgery within the past 5 years
  • Orthopedic disability that could impair movement or limit participation in the exercise protocol
  • Neurologic disorder that may affect neuromuscular function or gait (e.g., stroke, Parkinson's disease, peripheral neuropathy, traumatic brain injury)
  • Known metabolic or cardiovascular disease (hypertension alone is not exclusionary), or requiring medical clearance under ACSM guidelines without having obtained it
  • Metal implants in the lower extremities, cardiac pacemaker, or other implanted electronic device
  • Pregnancy
  • History of seizure disorder or fainting (syncope)

Study Plan

This section provides details of the study plan, including how the study is designed and what the study is measuring.

How is the study designed?

Design Details

  • Primary Purpose: Device Feasibility
  • Allocation: N/A
  • Interventional Model: Single Group Assignment
  • Masking: None (Open Label)

Arms and Interventions

Participant Group / Arm
Intervention / Treatment
Experimental: Hip Exoskeleton Walking

Healthy younger adults (18-39 years) and older adults (65+ years) walk on an instrumented treadmill under multiple conditions across 2-5 lab visits: no exoskeleton, wearing the CMU Hip Exoskeleton with zero torque, and with powered assistance from the AI-driven hip exoskeleton controller (assistance limited to <0.2 Nm/kg). The primary objective is to evaluate real-time gait phase estimation accuracy of the prototype controller across users; metabolic cost, biomechanics, and muscle activity are recorded as secondary measures.

A lightweight (3.5 kg) research hip exoskeleton developed at Carnegie Mellon University with powered hip flexion/extension joints, a backpack-mounted computer and battery, and thigh and waist interfaces secured with straps. An AI-driven controller estimates the user's gait state from joint angle and wearable sensor data and commands assistive hip torque in real time, limited to <0.2 Nm/kg (peak device torque 18 Nm). Participants walk on a treadmill with powered assistance, with the device worn at zero torque, and without the device. The device was determined by the UT Austin IRB to be a non-significant risk device.

Other Names:
  • CMU Hip Exoskeleton

What is the study measuring?

Primary Outcome Measures

Outcome Measure
Measure Description
Time Frame
Gait Phase Estimation Error of the AI-Driven Hip Exoskeleton Controller
Time Frame: Day 1

Root-mean-square error (RMSE) between the controller's real-time gait phase estimate and ground-truth gait phase derived from force-instrumented treadmill data during powered hip exoskeleton walking, reported as percent of the gait cycle (%). Used to evaluate feasibility of the prototype controller across younger and older adult users.

Day 1

Secondary Outcome Measures

Outcome Measure
Measure Description
Time Frame
Net Metabolic Power During Powered Hip Exoskeleton Walking
Time Frame: Day 1

Net metabolic power (W/kg), measured by open-circuit indirect calorimetry with resting value subtracted, during treadmill walking at 1.25 m/s with powered hip exoskeleton assistance.

Day 1
Peak Hip Joint Moment During Powered Hip Exoskeleton Walking
Time Frame: Day 1

Peak biological hip extension moment (Nm/kg), computed from motion capture and force-instrumented treadmill data, during treadmill walking with powered hip exoskeleton assistance.

Day 1
Lower-Limb Muscle Activity During Powered Hip Exoskeleton Walking
Time Frame: Day 1

Average electromyography (EMG) amplitude normalized to maximum voluntary contraction (%MVC), averaged across eight recorded lower-limb muscles, during treadmill walking with powered hip exoskeleton assistance.

Day 1

Collaborators and Investigators

This is where you will find people and organizations involved with this study.

Sponsor

Collaborators

Investigators

  • Principal Investigator: Inseung Kang, PhD, Carnegie Mellon University

Study record dates

These dates track the progress of study record and summary results submissions to ClinicalTrials.gov. Study records and reported results are reviewed by the National Library of Medicine (NLM) to make sure they meet specific quality control standards before being posted on the public website.

Study Major Dates

Study Start (Actual)

August 6, 2026

Primary Completion (Estimated)

August 31, 2027

Study Completion (Estimated)

January 31, 2028

Study Registration Dates

First Submitted

November 3, 2026

First Submitted That Met QC Criteria

November 10, 2026

First Posted (Actual)

November 16, 2026

Study Record Updates

Last Update Posted (Actual)

November 16, 2026

Last Update Submitted That Met QC Criteria

November 10, 2026

Last Verified

November 1, 2026

More Information

Terms related to this study

Other Study ID Numbers

  • STUDY2026_00000477
  • R21EB037268 (U.S. NIH Grant/Contract)

Plan for Individual participant data (IPD)

Plan to Share Individual Participant Data (IPD)?

YES

IPD Plan Description

De-identified individual participant data underlying published results will be shared, including biomechanics (motion capture, ground reaction forces), electromyography, metabolic, and exoskeleton sensor/controller data, along with participant characteristics (age group, sex, height, mass). Identifiable information, including video recordings, will not be shared.

IPD Sharing Time Frame

Data will be made available no later than the time of associated publication or the end of the award period, whichever comes first, and will remain available for at least 5 years.

IPD Sharing Access Criteria

De-identified data and supporting information will be deposited in a public data repository and made openly available to researchers without restriction, subject to the repository's terms of use.

IPD Sharing Supporting Information Type

  • STUDY_PROTOCOL
  • ANALYTIC_CODE

Drug and device information, study documents

Studies a U.S. FDA-regulated drug product

No

Studies a U.S. FDA-regulated device product

Yes

product manufactured in and exported from the U.S.

No

This information was retrieved directly from the website clinicaltrials.gov without any changes. If you have any requests to change, remove or update your study details, please contact register@clinicaltrials.gov. As soon as a change is implemented on clinicaltrials.gov, this will be updated automatically on our website as well.