ASTM F3578-22

Standard Test Method for Evaluating Exoskeleton Fall Risk due to Stumbling

SIGNIFICANCE AND USE
5.1 There is strong evidence that exoskeletons can physically augment and assist users. They are typically designed and optimized with specific tasks in mind and initially tested in controlled lab or field settings. However, in the real world exoskeletons encounter less structured environments and situations (for example, hospital rooms, factory floors, construction sites, or even personal homes). In order to accelerate the adoption of exoskeletons in society, understanding their safety in the presence of perturbations is helpful. The testing results of the exoskeleton shall describe the extent to which the exoskeleton improves, inhibits, or maintains a user’s ability to recover from stumbles, thus providing exoskeleton wearers and prescribers (for example, patients, clinicians, industry leaders, factory workers) with additional information about device performance and expectations.  
5.2 The standard test apparatus and setup (see Section 6) is specified to be easily fabricated and implemented in gait or motion analysis laboratories. Variants of the apparatus, control algorithm, and test setup are acceptable to allow implementation in various lab settings with ranging experimental capabilities. The standard test setup and apparatus can also be used to support training and establish proficiency of exoskeleton users, as well as provide manufacturers with information about the performance of their exoskeleton(s) for tasks.
SCOPE
1.1 Purpose:  
1.1.1 The purpose of this test method is to evaluate the extent to which an exoskeleton (see Section 3) improves, inhibits, or maintains (that is, does not affect) a user’s ability to recover from a stumble perturbation.  
1.1.2 Exoskeletons are designed to assist specific tasks and initially tested in controlled lab or controlled field settings. However, in the real world exoskeletons encounter less structured environments and situations (for example, hospital rooms, factory floors, construction sites). Even without exoskeletons people will stumble (that is, trip) or scuff their foot. It would be helpful to understand how wearing an exoskeleton affects a person’s ability to recover from a stumble perturbation. Is one’s ability to recover hampered, enhanced, or unaltered when using an exoskeleton? This test method specifies test setup, procedure, and recording to standardize testing exoskeleton user stumble recovery.  
1.2 Performing Location—This test method shall be performed in a testing laboratory where the specified apparatus and environmental conditions are available and implemented.  
1.3 Units—The values stated in SI units are to be regarded as the standard. The values given in parentheses are not precise mathematical conversions to inch-pound units. They are close approximate equivalents for the purpose of specifying material dimensions or quantities that are readily available to avoid excessive fabrication costs of test apparatuses while maintaining repeatability and reproducibility of the test method results. These values given in parentheses are provided for information only and are not considered standard.  
1.4 This standard does not purport to address all of the safety concerns, if any, associated with its use. It is the responsibility of the user of this standard to establish appropriate safety, health, and environmental practices and determine the applicability of regulatory limitations prior to use.  
1.5 This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.

General Information

• Status: Published
• Publication Date: 14-Jun-2022
• ICS: 11.180.10 - Aids and adaptation for moving
• ICS: 25.040.30 - Industrial robots. Manipulators
• Technical Committee: F48 - Exoskeletons and Exosuits
• Drafting Committee: F48.02 - Human Factors and Ergonomics

Relations

• Refers: ASTM F3443-20 - Standard Practice for Load Handling When Using an Exoskeleton
• Effective Date: 15-Jun-2020
• Refers: ASTM F3427-20 - Standard Practice for Documenting Environmental Conditions for Utilization with Exoskeleton Test Methods
• Effective Date: 01-Mar-2020
• Refers: ASTM F3323-19a - Standard Terminology for Exoskeletons and Exosuits
• Effective Date: 01-Oct-2019
• Refers: ASTM F3323-19 - Standard Terminology for Exoskeletons and Exosuits
• Effective Date: 01-Apr-2019
• Refers: ASTM F3323-18 - Standard Terminology for Exoskeletons and Exosuits
• Effective Date: 01-Sep-2018

Overview

ASTM F3578-22: Standard Test Method for Evaluating Exoskeleton Fall Risk due to Stumbling provides a comprehensive framework for assessing how exoskeletons impact a user’s ability to recover from stumbling events. With exoskeletons increasingly used in diverse real-world environments-such as hospitals, factories, construction sites, and homes-this standard addresses a crucial safety concern by enabling consistent and rigorous evaluation of exoskeleton-assisted stumble recovery. The method facilitates standardized lab-based testing to determine if an exoskeleton improves, impedes, or has no effect on fall risk.

Key Topics

• Test Setup and Procedure
  The method defines a laboratory-based protocol using an apparatus to introduce controlled stumbles while an exoskeleton user walks on a treadmill. Safety harnesses, instrumented treadmills, and sensory occlusion equipment are mandated to ensure consistent results.

• Perturbation and Fall Metrics
  Stumble perturbations are induced by a weighted obstacle. The main metric for fall risk is the number of falls, as determined by safety harness data and objective indicators such as trunk flexion after a perturbation. These results are compared between baseline (no exoskeleton) and exoskeleton-assisted walking.

• Variability and Customization
  The standard allows for customization in factors such as walking speed, obstacle characteristics, the timing and side of perturbations, environmental simulated conditions (e.g., different surface types, load carrying), and test participant selection in order to represent real-world use.

• Safety and Reproducibility
  The standard mandates safety equipment, familiarization periods, and consistent documentation of environmental and exoskeleton configurations to maximize repeatability and reliability of results.

Applications

• Product Evaluation and Development
  Manufacturers can utilize ASTM F3578-22 to objectively assess and compare the stumble recovery performance of their exoskeletons. Data generated helps inform design improvements focused on safety and user mobility.

• Clinical Rehabilitation
  Clinicians and therapists can reference this standardized method when prescribing exoskeletons for patients, ensuring that chosen devices meet safety benchmarks for fall risk in variable environments.

• Occupational Health and Safety
  Industries employing exoskeletons to assist workers-such as manufacturing or construction-can apply this test method to better understand device performance in preventing falls, contributing to workplace injury prevention strategies.

• Training and Proficiency Assessment
  The test protocol supports training programs and proficiency validation for exoskeleton users by providing a controlled environment to experience and recover from stumbling events.

Related Standards

ASTM F3578-22 references several complementary standards important for exoskeleton testing and documentation, including:

• ASTM F3323: Terminology for Exoskeletons and Exosuits
  Provides standardized language for exoskeleton and exosuit-related terms.

• ASTM F3427: Practice for Documenting Environmental Conditions for Utilization with Exoskeleton Test Methods
  Outlines requirements for controlling and documenting environmental test variables.

• ASTM F3443: Practice for Load Handling When Using an Exoskeleton
  Addresses test protocols for evaluating exoskeletons under varying load conditions.

• ASTM F3474: Practice for Establishing Exoskeleton Functional Ergonomic Parameters and Test Metrics
  Defines approaches for measuring and evaluating ergonomic parameters in exoskeleton performance.

Practical Value

Implementing ASTM F3578-22 ensures a consistent, measurable assessment of how exoskeletons affect user stability during stumbling incidents. This reliability not only benefits product developers and clinicians in their decision-making but also supports user confidence and regulatory compliance. As the exoskeleton market grows, adopting standardized safety test methods like ASTM F3578-22 is vital for safeguarding users and promoting innovation.

Keywords: exoskeleton safety, fall risk, stumble recovery, standardized test method, occupational exoskeletons, clinical exoskeleton evaluation, ASTM F3578, exosuit testing, rehabilitation devices, gait analysis.

Standards Content (Sample)

This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the
Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.
Designation: F3578 − 22
Standard Test Method for
Evaluating Exoskeleton Fall Risk due to Stumbling
This standard is issued under the fixed designation F3578; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.
1. Scope ization established in the Decision on Principles for the
Development of International Standards, Guides and Recom-
1.1 Purpose:
mendations issued by the World Trade Organization Technical
1.1.1 The purpose of this test method is to evaluate the
Barriers to Trade (TBT) Committee.
extent to which an exoskeleton (see Section 3) improves,
inhibits, or maintains (that is, does not affect) a user’s ability to
2. Referenced Documents
recover from a stumble perturbation.
2.1 ASTM Standards:
1.1.2 Exoskeletons are designed to assist specific tasks and
F3323 Terminology for Exoskeletons and Exosuits
initially tested in controlled lab or controlled field settings.
F3427 Practice for Documenting Environmental Conditions
However, in the real world exoskeletons encounter less struc-
for Utilization with Exoskeleton Test Methods
tured environments and situations (for example, hospital
F3443 Practice for Load Handling When Using an Exoskel-
rooms, factory floors, construction sites). Even without exo-
eton
skeletons people will stumble (that is, trip) or scuff their foot.
F3474 Practice for Establishing Exoskeleton Functional Er-
It would be helpful to understand how wearing an exoskeleton
gonomic Parameters and Test Metrics
affects a person’s ability to recover from a stumble perturba-
tion. Is one’s ability to recover hampered, enhanced, or
3. Terminology
unaltered when using an exoskeleton? This test method speci-
3.1 General terminology for ASTM Committee F48 stan-
fies test setup, procedure, and recording to standardize testing
dards is listed in Terminology F3323. Terminology specific to
exoskeleton user stumble recovery.
this standard are shown in this section.
1.2 Performing Location—This test method shall be per-
3.2 Definitions:
formed in a testing laboratory where the specified apparatus
3.2.1 apparatus, n—a structure, object, test component, or
and environmental conditions are available and implemented.
artifactthereof,foundorplacedinanenvironmentandusedfor
1.3 Units—The values stated in SI units are to be regarded
a test.
as the standard.The values given in parentheses are not precise
3.2.2 artifact, n—a representative of real structure(s),
mathematical conversions to inch-pound units. They are close
object(s), or test component(s) and used for a test.
approximate equivalents for the purpose of specifying material
3.2.3 perturbation, n—external disruption to body move-
dimensions or quantities that are readily available to avoid
excessive fabrication costs of test apparatuses while maintain- ment; in this standard, a perturbation specifically refers to a
disruption to the lower limb trajectory during swing phase of
ing repeatability and reproducibility of the test method results.
These values given in parentheses are provided for information gait due to an obstacle.
only and are not considered standard.
3.2.4 scuff, n—perturbation to the bottom of the foot during
1.4 This standard does not purport to address all of the swing phase.
safety concerns, if any, associated with its use. It is the
3.2.5 stumble, n—the act of tripping or losing balance as a
responsibility of the user of this standard to establish appro-
result of a perturbation to the front of the foot/toe during swing
priate safety, health, and environmental practices and deter-
phase.
mine the applicability of regulatory limitations prior to use.
4. Summary of Test Method
1.5 This international standard was developed in accor-
dance with internationally recognized principles on standard-
4.1 The task for this test method, exoskeleton user stumble
recovery, is defined as the exoskeleton user stumbling during
This test method is under the jurisdiction of ASTM Committee F48 on
Exoskeletons and Exosuits and is the direct responsibility of Subcommittee F48.02 For referenced ASTM standards, visit the ASTM website, www.astm.org, or
on Human Factors and Ergonomics. contact ASTM Customer Service at [e-mail]. For Annual Book of ASTM
Current edition approved June 15, 2022. Published July 2022. DOI: 10.1520/ Standards volume information, refer to the standard’s Document Summary page on
F3578-22. the ASTM website.
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States
F3578 − 22
walking. This task is used to assess the extent to which an 4.6 Once the test begins, there shall be no verbal commu-
exoskeleton improves, inhibits, or maintains a person’s ability nication between the exoskeleton user and the test supervisor
to recover from a stumble. The task involves an exoskeleton regarding the performance of a test repetition, other than
user walking on a treadmill and experiencing a perturbation. instructions on when to start, notifications of faults, and any
safety concerns or physical discomfort. The user shall have the
4.1.1 This test method can also be used to examine the
authority to request that the test be stopped at any point during
response of the exoskeleton user to scuffs and other perturba-
the trial. However, it is the test supervisor’s authority to judge
tion types, see 6.5.3.
the completeness of the repetition. If the testing is stopped
4.2 The number of perturbations and conditions for each
before all specified trials are completed, the test is marked as
perturbation will be defined by the test requester prior to the
incomplete.
test. The recommended apparatuses are described in Section 5.
4.7 The test requester has the authority to select the param-
The test requester also selects the participants for the test,
eters that may affect the user for the task. The test requester
which should be representative of the expected user population
also has the authority to select test methods that constitute the
of the exoskeleton to be tested.
test event, to select one or more test site(s) at which the test
4.3 The exoskeleton’s capability is defined as the fall or methods are implemented, to determine the corresponding
recovery outcome of the task, as determined by the weight- statistical reliability and confidence levels of the results for
bearing assistance provided by a force-instrumented overhead eachofthetestmethods,andtoestablishtheparticipationrules
safety harness (see Section 6) and optionally the maximum including the testing schedules and the test environmental
trunk flexion after the perturbation as measured, for instance, conditions. As such, variations to this test method are also
described in this standard, including:
by optical motion capture, a goniometer, or inertial measure-
ment unit (IMU). These outcome metrics are evaluated by 4.7.1 Leg swing percentage of perturbation;
4.7.2 Side experiencing perturbation (that is, right limb vs.
comparing the baseline case (that is, no exoskeleton) to the
exoskeleton case to determine whether the exoskeleton left limb);
4.7.3 Obstacle weight and dimensions:
improves, impedes, or maintains the exoskeleton user’s perfor-
mance(thatis,fewerfallsorlesstrunkflexion,orboth,relative 4.7.3.1 Changing the dimensions of the obstacle can allow
the test to include scuffs and other perturbations, see 6.5.3.
to baseline are improvements; more falls or more trunk flexion,
4.7.4 Walking speed;
or both, are impediments). The test requester can specify the
4.7.5 Environmental conditions including, for example,
number of perturbations, and as such, total falls out of total
ground surfaces that are level or sloped or uneven/undulating;
tests shall be recorded and used as the exoskeleton’s fall risk,
hard or soft; temperatures levels that are normal or extreme
and can be compared to the user’s fall risk in the baseline case,
(Practice F3427); and
see 8.4.
4.7.6 While carrying a load(s) (Practice F3443).
4.3.1 All users are required to wear a harness to catch them
after a fall to mitigate risk of injury. It is suggested that if the
5. Significance and Use
load cell measures >50 % bodyweight after a perturbation trial,
5.1 There is strong evidence that exoskeletons can physi-
even those including an apparent recovery, that trial should be
cally augment and assist users.They are typically designed and
recorded as a fall.
optimized with specific tasks in mind and initially tested in
4.4 The exoskeleton user may undergo a familiarization
controlled lab or field settings. However, in the real world
period and practice walking on the treadmill with the sensory
exoskeletons encounter less structured environments and situ-
occlusion equipment (that is, dribble goggles to block the
ations (for example, hospital rooms, factory floors, construc-
inferior visual field and wireless earbuds playing white noise
tion sites, or even personal homes). In order to accelerate the
along with noise cancelling headphones to block hearing, see
adoption of exoskeletons in society, understanding their safety
Section 6) and harness prior to the test, see 9.8.8.
inthepresenceofperturbationsishelpful.Thetestingresultsof
4.4.1 Practice is encouraged due to the potentially disori-
the exoskeleton shall describe the extent to which the exoskel-
enting initial effects of the sensory occlusion equipment or
eton improves, inhibits, or maintains a user’s ability to recover
discomfort associated with wearing a harness, or both. It is
from stumbles, thus providing exoskeleton wearers and pre-
suggested that the exoskeleton user practice walking with and
scribers (for example, patients, clinicians, industry leaders,
without the exoskeleton until they indicate they are comfort-
factory workers) with additional information about device
able walking continuously for 120 s or more.The practice time
performance and expectations.
and protocol, as well as any prior experience by the user
5.2 The standard test apparatus and setup (see Section 6)is
wearing the exoskeleton, should be described in the test report.
specified to be easily fabricated and implemented in gait or
4.5 Additionally, acclimation to the perturbation protocol is motion analysis laboratories. Variants of the apparatus, control
recommended by allowing the user to experience at least one algorithm, and test setup are acceptable to allow implementa-
perturbation for both exoskeleton and baseline (no exoskel- tion in various lab settings with ranging experimental capabili-
eton) cases to avoid inclusion of a potentially inconsistent ties. The standard test setup and apparatus can also be used to
first-time response in the data, see 9.8.8. The acclimation time support training and establish proficiency of exoskeleton users,
and protocol, as well as any prior experience by the user with as well as provide manufacturers with information about the
the test apparatus, should be included in the test report. performance of their exoskeleton(s) for tasks.
F3578 − 22
6. Apparatus 6.2 Optional Electronics and Algorithm:
6.2.1 A predictive targeting algorithm can be used to elicit
6.1 Recommended Apparatus:
precisely timed perturbations during a given stride via com-
6.1.1 A weighted obstacle must be introduced to the tread-
puter assistance. This algorithm is freely available online.
mill belt to induce the perturbation, in such a way that the
6.2.2 Force and moment data from an instrumented tread-
exoskeleton user is unaware of the obstacle’s approach (that is,
mill are required for the use of the algorithm.
low impulse and sound). This can be done more challengingly
by hand, or through the use of an obstacle delivery apparatus,
6.2.3 Data is low-pass filtered to remove noise and clipped
one suggested example of which is provided here. to prevent excessive voltage signals from damaging the micro-
6.1.2 The ramp consists of an acrylic track attached to an
processorbeforeitundergoesanalogtodigitalconversiontobe
aluminum frame with adjustable, vibration-damping feet. The inputs for the algorithm which runs on a microprocessor.
obstacle is held at a given point along the ramp via an
6.2.4 The algorithm uses the cyclic nature of gait, along
electromagnet, which is held by a rod located by a pair of holes
with known time and position constants of the apparatus, to
in the ramp (Fig. 1). The obstacle sits on the track via a set of
determine when the obstacle must be released to achieve a
flangedrollerbearingsmountedonshoulderboltsthreadedinto
perturbation at a specific time. The algorithm computes key
each corner (Fig. 1). The end of the ramp is parallel to the
stride time metrics via gait event detection from the force and
surface of the treadmill and overhangs the front edge to
moment data inputs. Known apparatus time constants are
securelydeposittheobstacleontothetreadmill(Fig.2)without
preemptively inputted into the algorithm.
causingvibrationsoranimpactthatwouldalerttheparticipants
6.2.4.1 This algorithm requires a split-belt treadmill (that is,
to the impending perturbation. Bill of materials and instruc-
to collect force and moment data under each foot separately).
tions for building this ramp are freely available online. Note
6.3 Required Equipment:
that a large, padded bin is placed at the back end of the
treadmill to catch the obstacle. 6.3.1 Measurement Device—A load cell in-line with the
overhead harness will be provided to measure the bodyweight
(force) assisted following the perturbation.
6.3.1.1 It is suggested that if the load cell measures >50 %
S. T. King, M. E. Eveld, A. Martínez, K. E. Zelik, and M. Goldfarb, “A novel
bodyweight after a perturbation trial, even those including an
system for introducing precisely-controlled, unanticipated gait perturbations for the
study of stumble recovery,” J. NeuroEngineering Rehabil., vol. 16, no. 1, p. 69, Jun. apparent recovery, that trial should be recorded as a fall.
2019, doi: 10.1186/s12984-019-0527-7.
Asteelblock(1)restsonanacrylictrack(2)viaflangedbearingstacks(3).Theblockisheldinplacebyanelectromagnet(4),whosepositionisdeterminedbytheheight
of the metal rod (5). The track is mounted to an aluminum frame (6) with adjustable, vibration-damping feet (7). Foam (8) is adhered to the front and bottom of the block
to protect the exoskeleton user’s toes and reduce the impulsive loading on the treadmill, respectively.
FIG. 1 Obstacle Delivery Apparatus
F3578 − 22
The exoskeleton user walks on the instrumented treadmill. Ground reaction forces and moments are collected (1) and used to calculate the center of pressure under
the foot, which is then used to detect gait events. These gait events are used to calculate the time at which the obstacle should be released using the predictive targeting
algorithm (2). At this time the electromagnet turns off (3) and releases the obstacle onto the treadmill such that a perturbation is introduced (4) at the desired percent of
swing phase.
FIG. 2 Schematic of the Stumble Perturbation System
3-6
6.3.2 Safety Equipment, including: a harness to prevent the late. Prior studies have found that individual responses differ
exoskeleton user from falling onto the treadmill following a depending on when in swing phase the perturbations occur.
perturbation; and protective footwear to avoid direct impact of 6.5.2 Side of Perturbation (that is, left vs. right limb)—The
the obstacle to the exoskeleton user’s toes. apparatus shown in Fig. 1 can be reproduced in order to
6.3.3 Testing Equipment, including: dribble goggles to performtheperturbationoneitherlimboftheexoskeletonuser.
block the exoskeleton user’s inferior visual field and prevent 6.5.2.1 It is recommended that perturbations should be
them from observing the approaching obstacle; wireless ear- performed on both limbs in order to (1) prevent anticipation or
buds to play white noise and prevent the exoskeleton user from (2) to account for potentially differing responses due to
hearing the release of the obstacle; and passive, noise- participant/exoskeleton asymmetries, or both. First, for healthy
cancelling, protective headphones to further prevent the exo- adults, limb dominance has not been shown to play a role in
skeleton user from hearing the release of the obstacle. fall likelihood or recovery strategy selection, as it is primarily
a reflexive response. Thus, for cases of healthy adults wearing
6.4 Optional Equipment:
a symmetrically weighted and controlled exoskeleton, which
6.4.1 Safety Equipment—For example: heart rate monitor,
limb is tripped (that is, left versus right) should not influence
pulse oximeter, or any other safety equipment which the test
fall outcomes. However, randomizing which limb is tripped
requester deems necessary can be at the ready as needed.
has been recommended in order to prevent any anticipation
6.4.2 User Measurement Devices—For example: motion
from the participant of the nature of when/where the perturba-
capture system, goniometer, IMU, muscle activity sensors,
tion will occur. Second, if the exoskeleton is unilateral or
heart rate monitors.
6.5 The test sponsor has the authority to customize the
A. M. Schillings, B. M. Van Wezel, and J. Duysens, “Mechanically induced
following in order to incorporate various test conditions:
stumbling during human treadmill walking,” J. Neurosci. Methods, vol. 67, no. 1,
6.5.1 Swing Percentage of Perturbation—Using the op-
pp. 11–17, Jul. 1996.
tional predictive algorithm or by manual release, the timing of
J. J. Eng, D. A. Winter, and A. E. Patla, “Strategies for recovery from a trip in
early and late swing during human walking,” Exp. Brain Res., vol. 102, no. 2, pp.
the perturbation (that is, when in swing phase the perturbation
339–349, Dec. 1994, doi: 10.1007/BF00227520.
occurs) can be controlled.
C. Shirota,A
...