Chapter 4

Exploring user requirements for a lower body

soft exoskeleton to assist mobility
Valerie Power1, Adam de Eyto1, Christoph Bauer 2,
Corien Nikamp3, Samuel Schülein4, Jeanette Müller5,
Jesús Ortiz6, and Leonard O’Sullivan1

Abstract
Understanding the requirements of potential users is crucial to the successful design of
wearable exoskeleton systems. Considering user requirements throughout the design
process optimises the likelihood of user uptake and facilitates adherence to the use of
wearable systems. This chapter describes the application of a user-centred design
approach to the development of a soft lower body assistive exoskeleton for individuals
with mild to moderate mobility impairment. Examples of the identification and
characterisation of user groups, the use of qualitative and quantitative research
methods to explore user requirements, and the implications of user requirements for
soft exoskeleton technologies are presented.

Keywords: User requirements; user-centred design; soft robotics; assistive

exoskeleton; gait; mobility

4.1 Introduction

Designing a successful wearable exoskeleton system requires extensive technical

expertise and high levels of innovation. Yet, even the most advanced, innovative
systems cannot be considered successful should they fail to be accepted by their
intended users.

1
School of Design, University of Limerick, Ireland
2
Institut für Physiotherapie, ZHAW Zürcher Hochschule für Angewandte Wissenschaften, Switzerland
3
Roessingh Research and Development, Netherlands
4
Geriatrie-Zentrum Erlangen, Germany
5
accelopment AG, Switzerland
6
Department of Advanced Robotics, Istituto Italiano di Tecnologia, Italy
68 Wearable exoskeleton systems: design, control and applications

Technology acceptance has been studied across a variety of technologies, users

and contexts of use, particularly over the past three decades, as technology has become
increasingly pervasive in most facets of society. The Technology Acceptance Model –
and its variants – is one of the most commonly-used frameworks for understanding user
acceptance of technology [1–3]. Technology acceptance suggests that usage behaviour
is strongly influenced by the user’s perceived usefulness and perceived ease of use of a
given technology, along with social factors. The challenges of designing highly
complex wearable exoskeleton systems which users perceive as being both useful and
easy to use are not insignificant and hence need to be carefully addressed.
The application of wearable exoskeleton systems as assistive technologies pre-
sents many opportunities, but also poses additional challenges. Assistive technology
is considered from a ‘person-first’ approach, where the person, his/her occupation/
activities and environment take precedence, with the technology adapting to indivi-
dual needs e.g. the Human Activity Assistive Technology model [4]. To design
wearable assistive exoskeletons that are fit for purpose, collaboration is required
across multiple disciplines, including (but not limited to) technology, design,
healthcare, psychology and industry.
This chapter will provide insights into the process of designing wearable
assistive exoskeleton systems from a user-centred perspective. An example of the
design of a soft lower body exoskeleton to assist mobility will be used throughout
the chapter to present the principles, methods, opportunities and challenges asso-
ciated with the user-centred design of assistive exoskeleton systems in a practical
context. Since the focus of this chapter is on user-centred design, the user
requirements discussed will be related to users and their needs, rather than technical
specifications. Specifically, this chapter will present:
● Principles of user-centred design;
● An overview of the XoSoft project aimed at developing a lower body soft
exoskeleton to assist human mobility;
● Identification of user groups for a lower body soft assistive exoskeleton;
● An example of mixed methods research to examine the requirements of
potential users of a soft assistive exoskeleton for human mobility;
● The implications for soft exoskeleton technologies posed by user requirements.

4.2 User-centred design

When designing and developing products, particularly assistive devices such as
XoSoft, it is recommended to involve both primary end users and secondary pro-
fessional users in the design and development process. This helps to ensure that
users’ personal expectations are met, insofar as is possible, thereby maximising the
possibilities for user uptake and acceptance [5]. To formulate a clear design brief,
it is also necessary to explore and understand the demographic factors and
experiences of users which contribute to the formation of their expectations and
requirements. Observational design research methodologies, ethnographic research,
clinical assessment and elements of co-design can be used or combined in mixed
 Exploring user requirements 69

methods approaches, to enhance the early stage user involvement in the design
process. These methods help to gather rigorous user feedback and rich contextual
information on which to base the design specification [6].
In the early stages of development, a user-led approach ensures that the ‘voice
of the user’ is both listened to and advocated for while developing the technological
specification of the ‘product’ and the ‘product service system’ (PSS). The identi-
fication of real users’ needs is an essential component to successfully adopting a
user-centred approach – often small omissions or oversights in the development
process can present the biggest challenges to primary and secondary users (SUs)
when the product reaches the market. Using the time and resources on following a
user-centred design approach at the beginning of a development process has been
proven to be highly beneficial in avoiding costly mistakes and the need for
re-designing in the later stages of development.
A user-centred approach to design is also recommended during the iterative
phases of development and prototyping. This can be achieved by investigating
if the subsequent prototype designs match the user requirements that were for-
mulated at the beginning of the project, via a user requirements validation study.
This approach allows developers to identify priorities for subsequent iteration
cycles. Seeking user testing, feedback and response to early stage prototypes
greatly enhances the chances of the product and PSS being relevant and successful
within the marketplace but more importantly it verifies that the product is fit for
purpose [7]. Highly regulated robotic assistive devices such as XoSoft require as
part of regulatory approval to include user feedback and validation as formal
elements of the design and development process.
SU feedback is also useful as a means of establishing what support and remote
assistance clinicians, carers and personal assistants can give through the product’s
life cycle. In the case of assistive robotic devices, the technical nature of the pro-
duct means that setup, customisation and maintenance of a product may be carried
out by SUs. SU needs are often as critical as those of primary users (PUs), and the
determination to invest in a technology can often weigh heavily on the ease of use
from a SU’s perspective.

4.3 XoSoft: a soft lower body exoskeleton to assist mobility

In Europe, it is estimated that over 51 million people experience a longstanding

difficulty with walking [8]. People who experience mobility limitations can use a
variety of assistive devices e.g. walkers, canes, manual and powered wheelchairs.
However, currently available assistive devices have many limitations and fre-
quently go unused, with up to 50% of users abandoning their devices soon after
receiving them [9]. Reasons for this include excessive bulkiness of devices,
usability issues with devices which occupy and/or load the upper limbs and the
significant attentional demands which some devices place on the user. Such pro-
blems not only limit feasibility for everyday use, but can also lead to adverse
effects, including falls, upper limb pain and potential injury [9–12]. Furthermore,
70 Wearable exoskeleton systems: design, control and applications

Communication and connected health

Central processing hub

Data gathering
Actuator control
Communication
Power

Soft sensing and actuation

Variable stiffness joint
Multi-joint actuation
Smart soft mechanical sensing
Inertial sensors (for 3D kinematics)

Pressure sensing
Eoin White, www.designfactors.ie, University of Limerick

Figure 4.1 Image of the XoSoft concept and its components

many assistive devices are inappropriately prescribed and/or not prescribed by

healthcare professionals, and users do not receive adequate training on safe and
appropriate usage [9].
An international, multidisciplinary consortium of researchers set out to
develop a soft, modular, lower limb exoskeleton (XoSoft) for older adults and
people with disabilities. Ultimately, XoSoft’s purpose is to help increase users’
mobility, and thus improve their health and quality of life.
XoSoft was developed by a consortium of nine partners comprising clinical
and research centres, universities, SMEs and an industrial partner with experience
in developing exoskeletons/assistive devices. User-centred design was a core ele-
ment of the XoSoft project. An iterative design process was undertaken, whereby
user requirements drove technical innovations and user feedback informed cycles
of design and prototype development.
Figure 4.1 displays an initial impression of the XoSoft concept and its key
features. Advanced textiles and smart materials are employed to create sensing
variable stiffness joints. Built-in sensors communicate the user’s motion and
intention to the controlling unit for analysis, to determine and provide the appro-
priate level of assistance, by way of actuators. Depending on the user’s need at a
given moment, the device will provide support, release or freedom of movement.
Specific details of the sensing and actuation technologies used are not available,
since XoSoft was in a conceptual stage of development at the time of writing.
 Exploring user requirements 71

4.4 Identifying users of a soft exoskeleton to assist mobility

In user-centred design of exoskeletons, it is crucial to clearly identify and under-
stand the target users, their characteristics and their needs. The aim is to design an
exoskeleton that is both functional and acceptable to users, and to ensure this,
a variety of data should be gathered from multiple sources e.g. demographic data on
the target user groups, user opinions on necessary and desired design features, data
on the context in which the exoskeleton will be used, user anthropometrics and
human motion data.
Users of a product may be categorised as primary, secondary and tertiary.
The following sections describe these user groups, using the target users of XoSoft
as examples of the populations who may be included each user group, and the
reasoning used to identify and select them. This is not an exhaustive list of the
potential users of a soft exoskeleton to assist mobility; the target users will vary
depending on the functions of the exoskeleton (e.g. provision of assistance, mon-
itoring and feedback on performance), its design (e.g. upper body, lower body,
single or multi-joint) and its context of use (e.g. clinical, industrial, etc.).

4.4.1 Primary users

PUs are the people who actually use and directly benefit from the exoskeleton.
XoSoft’s PUs are adults who require low to moderate levels of assistance with
mobility. Specifically, the XoSoft project focused on three distinct PU groups:
1. People with stroke;
2. People with incomplete spinal cord injury (SCI);
3. Older adults with mild–moderate mobility impairments.
These groups were selected using reasoning based on evidence of the opportunities,
challenges and practicalities of creating an exoskeleton for these specific
populations, as well as epidemiological data to identify potential global market sizes.
Examples of this evidence-based reasoning are provided in the following sub-sections.

4.4.1.1 Stroke
In 2010, there were 33 million stroke cases worldwide, with 16.9 million people
having a first stroke [13]. The nature and severity of post-stroke impairments vary
between individuals, depending on the type and severity of stroke. Motor deficits are
the most highly prevalent post-stroke impairment, commonly presenting as weakness
affecting one side of the body (hemiparesis); approximately 72% of people experience
lower limb motor deficits and 77% experience upper limb motor deficits [14].
Compared to healthy individuals, people who have experienced stroke gen-
erally exhibit slower, more asymmetrical gait [15]. A common pattern of motor
deficits observed in post-stroke gait includes decreased ankle dorsiflexion [16–19],
decreased knee flexion [18–20] and hip flexion [15] during the swing phases of
gait, which combine to produce insufficient foot clearance and increased risk
of trips and falls. Various compensatory movement strategies around the pelvis
72 Wearable exoskeleton systems: design, control and applications

[20–22] and hip [17,23] may be adopted to overcome the effects of these deficits,
but such compensatory strategies can greatly increase the energy costs of walking
[15]. Additionally, impairments related to sensory systems (e.g. vision, vestibular,
somatosensory) and sensory organisation can further compromise gait in the stroke
population [24,25].
Approximately one-third of community-dwelling people with stroke require
supervision for community ambulation [26] indicating potential opportunities
for assistive exoskeletons in this sector. Sensing and actuation technologies may
be used to provide alternative sources of feedback to the user, as well as assistive
lower limb joint stiffness/torques, which may offer valuable support during
activities of daily living.
However, a number of challenges exist. The varying types and severities of
clinical presentations following stroke present a design challenge, since design
features required by some users may not be useful – or may present barriers to use –
for others. Concomitant upper limb impairment may limit the ability to use devices
to assist mobility, particularly exoskeletons which must be donned and doffed.
Additionally, altered muscle tone and impaired voluntary movement following
stroke can present challenges to the control of wearable devices like XoSoft.
Furthermore, cognitive impairment has been observed in 35%–38% of individuals
3 months post-stroke and is reported to be even more prevalent in acute stages
and among those with greater levels of disability [27,28]. Memory, orientation,
language and attention are the cognitive domains most likely to be affected [28],
which presents further challenges to usability and may increase the risk of inap-
propriate use of a wearable assistive exoskeleton.

4.4.1.2 Incomplete SCI

SCI may be categorised according to its aetiology: traumatic (T-SCI) or non-
traumatic (NT-SCI). Motor and sensory impairments are experienced by both groups,
thus wearable assistive devices offer great potential to assist all. However, the nature
and extent of these impairments vary depending on the location and extent of the spinal
cord lesion, therefore the potential requirements for assistance are diverse.
The age-adjusted incidence for T-SCI in Europe ranges from 8 to 131 (median
20.1) per million persons per year. The prevalence in Europe ranges from 227 to
526 (median 351) per million population [29]. These statistics concern both com-
plete and incomplete SCI. Approximately 45% of individuals with T-SCI experi-
ence an incomplete tetraplegia, 21% an incomplete paraplegia; the remainder
experience complete SCI [30]. The T-SCI population is generally young (average
age: 42 years) and approximately 80% are male [30]. Cognitive impairment is not a
feature of SCI, unless the trauma which led to the SCI also led to traumatic brain
injury, and therefore does not generally need to be factored into the design
requirements of this user group. In a European cohort, 8% of individuals of all
lesion types regained walking function with one or two leg braces 6 months after
injury [31]. Some rigid exoskeleton products are already available for individuals
with SCI (e.g. ReWalk, Eksobionics, HAL). These products are primarily for use
as rehabilitation tools at present, rather than as assistive devices, and although
 Exploring user requirements 73

evidence of their effects on mobility is still limited, some promising research is

beginning to emerge [32–34].
Reliable statistics concerning the incidence and prevalence of NT-SCI are less
readily available. It is a growing group and outnumbers T-SCI. Patients with NT-
SCI tend to be older and present with incomplete lesions [35,36]. Their changing
needs over time present a challenge to the successful design of wearable assistive
devices for this population, since NT-SCI or general health may deteriorate e.g.
people with NT-SCI due to tumours. Age-related complications – as described in
Section 4.4.1.3 – could also influence device usability in this group: for example,
although NT-SCI has no direct influence on cognitive function, age-related
cognitive impairment may be present in older adults with NT-SCI.

4.4.1.3 Older adults

The global population is ageing rapidly: the proportion of the world’s population
aged over 60 years is expected to rise from about 11% to 21% between 2000 and
2050, which represents an increase from 605 million to 2 billion older adults
worldwide [37]. Unfortunately, increasing life expectancy does not necessarily
equate to increasing healthy life expectancy. Multi-morbidity – living with two or
more chronic health conditions – is reported to affect over 60% of adults aged
65 years and over [38]. Living with multiple chronic conditions reduces life expec-
tancy, increases the risk of functional decline and reduces overall quality of life [39].
Mobility impairment is often a feature of age-related functional decline: data from
the Health and Retirement Study in the United States shows that 8.5% of adults aged 65
years and over experience mobility impairment to the extent that they have difficulty
walking across a room [40]. Disability is also becoming more prevalent among older
adults in high-income countries, where obesity, diabetes and cardiovascular disease are
widespread [41]. Given the prevalence of mobility impairments among older adults and
the ageing global population, requirements for assistance to maintain mobility will
continue to increase in the coming years, presenting considerable opportunities for
assistive devices, including assistive exoskeletons.
Frailty among older adults may be considered as an opportunity and a chal-
lenge to the use of a soft assistive exoskeleton. Frailty is often defined as a syn-
drome, which presents as age-associated decline in functioning across multiple
physiological systems [42]. It brings with it a high risk for falls, disability, hospi-
talisation and mortality. Frailty may be classified by the presence of three or more
of the following criteria: slow gait speed, low physical activity, unintentional
weight loss, exhaustion and muscle weakness. Older adults exhibiting one or two of
these criteria may be considered as ‘pre-frail’ [43]. While the declining mobility
seen with frailty suggests a potential need for assistive devices, exhaustion, general
weakness and the multiple co-morbidities seen in frail older populations may act as
barriers to uptake of, and adherence to, device usage.
An additional challenge to consider when designing for the older adult popu-
lation is age-related cognitive decline. In Europe, the prevalence of dementia in
adults aged 60–64 years is approximately 1%, but increases to almost 25% in those
aged 85 years and over [44]. Mild cognitive impairment can occur in up to 20% of
74 Wearable exoskeleton systems: design, control and applications

those aged 75 years and over [45]. This has clear implications for exoskeleton
usability; designing a complex system such that older users can interact with it in a
simple and intuitive manner is a significant design challenge.

4.4.2 Secondary users

SUs are those who have direct contact with PUs in professional or non-professional
capacities e.g. healthcare professionals, formal/informal caregivers, spouses, family
and friends. Professional SUs include healthcare professionals from multidisciplinary
backgrounds, such as medical doctors, nurses, physiotherapists, occupational thera-
pists, speech and language therapists, as well as professional care assistants, home
help service providers and other support staff. Professional SUs may also be grouped
according to the context in which they interact with PUs e.g. in hospitals, clinics,
rehabilitation units, private practices, in the home environment, etc.
Depending on the context in which SUs encounter PUs, their expectations and
requirements of a soft exoskeleton will differ. For example, professional SUs in
acute healthcare settings will be focused on the short-term usability and efficacy of
the device, since they often have very limited contact time with PUs. SUs in
community settings may place greater emphasis on long-term factors such as the
ability of the device to improve user safety, its robustness, and/or its adaptability to
PUs during long-term use. Examples of some of the main expectations of profes-
sional SUs in relation to an assistive device are summarised in Table 4.1. These
examples are based on previously published literature as well as expertise from
professional SUs.

Table 4.1 Examples of professional secondary users’ expectations in relation to a

soft assistive exoskeleton and related indicators of success

Expectations Indicators of success

Safety of device Device is appropriate for its intended purpose, has a low
incidence of adverse events, and evidence demonstrating
safety and efficacy is available [46]
Efficacy of device Improves clinical outcomes of PUs e.g. facilitates inde-
pendent mobility, reduces disability, halts/delays func-
tional decline, decreases the burden of care, maintains
PUs independence and do so in a cost-effective manner
[10,47,48]
Impact of device usage on Positive impacts on factors that influence overall quality of
PUs’ quality of life life, aside from health status, such as social relations and
participation in activities [49]
Accessibility of device to SUs Available to PUs and SUs at a feasible cost or via
and PUs appropriate reimbursement schemes
Alignment with professional Usage of the device is in keeping with the code of conduct
aims and standards of the relevant healthcare profession and its associated
professional body/regulatory authority in a given
jurisdiction
 Exploring user requirements 75

Non-professional SUs may include spouses, family members, friends, neighbours,

community and/or voluntary organisations and other informal caregivers. When
designing any assistive device for a person with impaired mobility, consideration must
be given to the opinions and experiences of such informal caregivers, particularly
spouses and immediate family members. Assistive devices – if adopted for long-term
use by PUs – become entwined in the everyday lives of these SUs, become part of their
shared home environments and can impart a sense of responsibility on SUs in relation
to the devices and their usage [50]. Caregivers can also play a role in encouraging
adoption of a device by PUs initially, can encourage, instruct and help with continued
use of a device and ultimately promote user satisfaction with their device [51].

4.4.3 Tertiary users

Tertiary users (TUs) are defined as individuals and/or organisations that do not
directly interact with the device, but have some influence on organising or enabling it
e.g. involvement in regulatory frameworks and/or making financial provisions for
devices. TUs are therefore often not considered as being part of the main focus when
developing a new device, but nonetheless they need to be considered as stakeholders.
TUs’ interests (Box 4.1) can be similar to those of SUs, but may be approached
from a different perspective.

Box 4.1 Sample factors of interest to TUs in relation to a

medical device, such as a soft exoskeleton
● Safety of the device – From a TU perspective, devices on the market
must be safe to use, and the benefits of use must outweigh any potential
risks. For medical devices, this includes obtaining the appropriate certi-
fication, registrations and conforming to all regulatory requirements.
● Cost of the device – Direct costs of the device may be considered, but TUs
often focus on the costs relative to the benefits offered by the device.
● Additional health/social care costs due to the device – Devices that
add costs must display clear benefits. TU support for a device that
involves increased overall costs would be rare and would require a
highly convincing case that the new device is replacing an older practice/
technology with significant positive changes and fewer adverse effects.
● Reduced health/social care costs due to the device – TUs view reduced
costs very favourably and often expect to see evidence of such effects.
Cost reductions may be achieved in several ways:
– By directly replacing an existing device but at a lower cost;
– By hastening healing or rehabilitation, thus lowering the overall
treatment cost;
– By increasing user independence and reducing need for assistance,
thus reducing costs of care.
76 Wearable exoskeleton systems: design, control and applications

● Impact on functional ability of PU – If an assistive device enables

users to be active in society again, including working and paying taxes if
possible, provision of the device is in the interest of TUs, as well as PUs
and SUs. This is seen as a highly favourable outcome, as it lowers overall
health/social care costs.
● Impact on quality of life – TUs may be primarily concerned with effects
on quality of life of PUs at population levels, rather than for individual
PUs. Overall, this would most likely not be a primary outcome measure
from the TU point of view.
● Willingness to pay for the device by PUs/general population – The
TUs to whom this relates will depend on the health/social systems in
operation. In countries where there is high social involvement in pro-
viding assistive devices, political decisions are required, often informed
by cost–benefit analyses. In countries with less social contribution, TUs
in the insurance industry, in private healthcare services and others e.g.
resellers, will play a greater role.

TUs may not be directly involved in the design and development of a device,
however they may be consulted during design for advice on issues related to their
interests e.g. when testing new devices. The interests of TUs must be considered to
ensure the success of a proposed device at a later stage. TUs are a diverse group
with varied interests, and Table 4.2 contains examples of some key types of TUs,
their interests and some factors they may consider as indicators of success in the
design of a soft assistive exoskeleton.

4.5 A mixed methods study to explore users’ design

requirements
The XoSoft project adopted a user-centred design focus: user needs formed the
foundations of the concept and were the main drivers of the iterative design pro-
cess. The needs of PUs and SUs are particularly crucial at this early stage, as
discussed in Section 4.4, since these groups will be directly involved in using
an assistive device such as XoSoft. Placing the design requirements of these users
to the fore aims to ensure that the concept will not only meet users’ basic
requirements, but also their personal expectations, insofar as is possible, thereby
maximising the possibilities for uptake and acceptance [5].
To formulate a clear design brief for the XoSoft concept, it was deemed
necessary to conduct a study of user requirements. The aim of this study was to
explore the functional needs and design requirements of a selection of individuals
with mobility impairments (PUs), and those who have regular contact with them on
professional and non-professional bases (SUs), in relation to a lower body soft
exoskeleton to assist mobility.
 Exploring user requirements 77

Table 4.2 Examples of tertiary users (TUs) and their expectations and indicators
of success with regard to a soft assistive exoskeleton

TU Expectations Indicators of success

Regulatory Ensure that devices are safe to use Low incidence of adverse events
authorities for intended purpose through
legislation
Prevent marketing of unsafe
devices
Authorities as Provide best possible healthcare Keeping costs within budget
health care service on a fixed/limited budget while achieving
providers acceptable levels of care
Adequate cost–benefit ratio. Will
need evidence of outcomes
before accepting
device for reimbursement
Insurance com- Provide devices for their customers Adequate cost–benefit ratio
panies that offer benefits in relation to May define the level(s) of evi-
the device cost dence that would be needed to
accept a new device for
reimbursement
Standardisation Create ‘common rules’ and provide Standards are widely used in
bodies framework for comparing through the community
testing
Influencers Advocates for certain groups or Their point of view is reflected in
products e.g. senior citizen groups decisions from stakeholders
or patient organisations that decide on use and reim-
bursement
Resellers (distri-Selling the device Profitable distribution of device
butors)
Private hospitals Offering the device as a therapeutic Increase of overall profits and
and rehabilita- tool reputation of treatment centre
tion centres Training PUs and SUs in the use
of the device
Highlighting the innovative character
of their institution

To achieve this aim, the following objectives were adopted:

● To conduct structured interviews with a range of potential PUs and SUs of a
soft assistive exoskeleton;
● To collect information on users’ requirements for and previous experiences of
assistive devices for mobility;
● To obtain users’ opinions on a soft assistive lower body exoskeleton, by pre-
senting a visual impression of the XoSoft concept;
● To gather information on user characteristics which may contribute to their
requirements and expectations of a device to assist mobility.
78 Wearable exoskeleton systems: design, control and applications

The following sections describe a study that was undertaken at the outset of the
XoSoft project to achieve this aim and objectives. Details on the study methods are
provided in Section 4.5.1, while the study results – presented in Section 4.5.2 –
provide valuable insights into the opinions and requirements of a variety of
potential users of a lower body soft assistive exoskeleton.

4.5.1 Methods
A cross-sectional mixed methods study was carried out across four European coun-
tries: Germany, Ireland, the Netherlands and Switzerland. A one-to-one structured
interview was used to ascertain each user’s opinions on and requirements for assistive
devices for mobility, and to gather relevant background information to characterise
users, and to provide context to their opinions and requirements. All research was
conducted in accordance with local research ethics regulations, including obtaining
written informed consent from individuals prior to participation, as appropriate.

4.5.1.1 Participants
Individuals who represented potential PUs and SUs of XoSoft were invited to
participate in four European countries. The countries were selected because they
were represented within the XoSoft consortium. Participants were recruited using a
purposive sampling method initially, with additional snowball sampling if appro-
priate. Figure 4.2 illustrates the flow of participants through the study.
The PU participants included were:
● Older adults with moderate to high levels of mobility impairment recruited from
eligible inpatients at a geriatric centre (termed ‘frail older adults’ hereafter);
● Older adults with mild to moderate mobility impairments recruited via older
adults’ community groups (termed ‘pre-frail older adults’ hereafter);
● Participants with mobility impairments due to stroke and incomplete SCI
recruited from a database of previous stroke research participants and rehabi-
litation centre patients who had opted to receive invitations to take part in
future research.
The presence of moderate to severe cognitive impairment – defined as a Mini-Mental
State Examination score of less than 18 – was an exclusion criterion for PU partici-
pants, as this was considered to be a potential limiting factor to the individual’s
capacity to provide informed consent to participate and/or to complete the interview
successfully. Individuals with scores of 18–24 were not excluded, since mild cognitive
impairment is reasonably prevalent among the selected PU groups [45,52].
SU participants were sourced via personal and professional networks of the
XoSoft consortium and included:
● Healthcare professionals from multidisciplinary backgrounds who specialise in
working with frail older adults at a geriatric centre;
● Healthcare professionals and human movement scientists experienced in
working with people with stroke and SCI in both research and rehabilitation
settings;
 Exploring user requirements 79

Target countries
DE, IE, NL, CH

Purposive sampling
(+Snowball sampling)

Score <18 Mini-Mental State

Examination
Score >18

PU participants (n = 15)
SU participants (n = 26)
Excluded from participation Stroke/SCI (n = 6)
Professional (n = 24)
(n = 0) Frail older adults (n = 6)
Non-professional (n = 2)
Pre-frail older adults (n = 3)

Participant characteristics
Assessment tools
assessment

• Demographics (age, height,

Clinical

• Functional Ambulation
weight, gender)
Categories (FAC)
• Health conditions
• Gait speed
• Living situation
• Barthel index
• Home environment

• General activity limitations • Demographics (age, gender)

Structured interview

• Mobility limitations • Relationship to PU(s)

• Requirements for assistance • Experience of assistive
• Experience of assistive devices
devices • Important features of an
• Important features of an assistive device
assistive device • Familiarity with exoskeletons
• Opinions on XoSoft • Opinions on XoSoft

Figure 4.2 Flow chart illustrating participants and components of the mixed
methods study

● Allied health professionals with experience across multiple fields – neurology,

musculoskeletal, geriatrics – in a range of settings, from acute to long-term
care and in both public and private healthcare practice;
● Non-professional SUs, including spouses, family members and friends of PUs.

4.5.1.2 Primary user assessment and interview

As shown in Figure 4.2, PU participants completed a brief clinical assessment and a
structured interview. The clinical assessment gathered information to characterise
PUs and their levels of mobility and independence. The following standardised
clinical assessment tools were included:
● Mini-Mental State Examination: A measure of cognitive function rated by the
researcher and used to screen participants for potential cognitive impairment.
Maximum score is 30. A score of less than 18 was considered to indicate
moderate to severe cognitive impairment, and therefore ineligibility to parti-
cipate further in the study [53].
● Functional Ambulation Categories (FAC): A researcher-rated tick-box
scale used to categorise participants’ independence with regard to walking
ability. Comprised of six broad categories, ranging from zero (participant cannot
80 Wearable exoskeleton systems: design, control and applications

walk, or needs help from two or more people) to five (participant can walk
independently in any environment) [54]. Walking aid usage was recorded in
addition to FAC score, since the FAC does not take this into account.
● Gait speed: Participants’ usual gait speed recorded using either a 5 or 10 m
walk test, depending on the setting and space available. It is a robust con-
tinuous measure of gait performance, and normative data is available for a
variety of populations [55].
● Barthel Index: A questionnaire measuring independence in activities of daily
living. Ten items are rated, and scores range from 0 to 20, with higher scores
indicating greater levels of independence [56].
The structured interview schedule – implemented across all data collection sites –
included questions on the following topics:
1. General activity limitations experienced by participants at home and in the
community;
2. Activity limitations specifically related to mobility and participants’ percep-
tions of these;
3. Participants’ requirements for assistance with mobility, including assistive
devices, if used;
4. Important features of an assistive device for mobility, as perceived by the
participant;
5. Participants’ opinions on a visual impression of the XoSoft concept (Figure 4.1)
and their perspectives on perceived benefits and challenges to its usage.
Cues were included in the schedule for each topic so that all interviewers could
provide standard clarification of topics and/or prompt discussion among partici-
pants, as required. Interview data were recorded in the form of detailed field notes,
which summarised all points made by participants in relation to each question on
the interview schedule.

4.5.1.3 Secondary user interview

SUs completed a structured interview only (Figure 4.2). The SU structured interview
schedule differed slightly from that of PUs and included questions on the following:
1. The nature of participants’ relationship(s) to potential PU(s) of XoSoft;
2. Participants’ experiences of assistive devices for mobility;
3. Challenges encountered in relation to assistive devices for mobility;
4. Important features of an assistive device for mobility, as perceived by the
participant;
5. Participants’ familiarity with assistive exoskeletons and perceptions of usefulness;
6. Opinions on a visual impression of the XoSoft concept, as shown to PUs
(Figure 4.1).
SUs, who were healthcare professionals and were familiar with exoskeletons,
were also asked an additional question regarding the device(s) with which they
were familiar, if they had used any such devices in their treatment of patients, and
 Exploring user requirements 81

their experiences of doing so. Cues were also used to clarify topics and/or prompt
discussion, as required, similar to the PU interview.

4.5.1.4 Data analysis

For PU clinical assessments, assessment tools were scored as per guidelines and
summarised using descriptive statistics. Summaries of each participant’s data for
each question in the structured interview were created by the interviewers. These
data were then compiled for each PU and SU group by a single researcher, and
analysed using thematic analysis to identify the main themes in each user group for
each question. Themes across user groups were then broadly categorised as
● Needs for assistance and experiences of assistive devices;
● Design requirements for wearable assistive devices for mobility;
● Perspectives on a soft assistive exoskeleton concept.

4.5.2 Results
A total of 15 PUs were interviewed: 6 in Germany, 6 in the Netherlands, and 3 in
Ireland. An overview of the characteristics of the PUs interviewed is displayed in
Table 4.3, split according to the distinct PU groups of interest. Frail and pre-frail
older PUs are presented separately, since frail PUs were considerably older
than pre-frail PUs and exhibited poorer mobility and greater levels of functional
disability, which would be expected to influence interview findings.
In total, 26 SUs participated (18 female, 8 male): 2 in Ireland, 6 in the Neth-
erlands, 12 in Germany and 6 in Switzerland. There were 24 professional and 2
non-professional SUs. Professional SUs included 12 physiotherapists, 5 physicians,
3 occupational therapists, 2 nurses and 2 other support staff. Twelve listed ger-
iatrics as their main area of work, four primarily worked in stroke rehabilitation,
two in SCI and eight worked with the general population. Three of the professional
SUs also noted that they currently work in clinical health research involving older
adults and neurological populations. Professional SUs were based in a variety of

Table 4.3 Overview of PU participant characteristics

Stroke/SCI Frail older adults Pre-frail older adults

(n ¼ 6) (n ¼ 6) (n ¼ 3)
Age (years) 55 (42–64) 87.5 (82–96) 74 (71–75)
Gender (m/f) 4/2 2/4 0/3
Height (m) 1.79 (1.70–1.96) 1.57 (1.50–1.66) 1.65 (1.60–1.68)
Weight (kg) 76.0 (63.0–100.0) 71.6 (48.2–78.2) 76.2 (76.0–88.9)
Gait Speed (m/s) 0.70 (0.34)* 0.61 (0.34–0.79) 1.40 (1.26–1.53)
FAC (score/5) 5 (5–5) 4 (3–4) 4 (3–5)
Barthel index (score/20) 19 (17–20) 12.5 (12–17) 19 (18–20)

*Gait speed of PUs with stroke is presented as mean (standard deviation).

Data are presented as median (minimum–maximum). SCI, spinal cord injury; FAC, functional ambu-
lation categories.
82 Wearable exoskeleton systems: design, control and applications

settings: 12 in a specialist geriatric centre, 5 in rehabilitation centres, 4 in private/

general practice, 2 in a hospital and 1 in a nursing home. Professional SUs had a
mean age of 37.1  9.0 and 12.0  8.3 years of experience. Non-professional
SUs included one spouse and one close friend of a PU.

4.5.2.1 Existing needs for and experiences of assistive devices

Although PUs with stroke and SCI were all independently mobile (FAC scores ¼ 5,
Table 4.3), the impairments of body structures and functions which they experienced
were not uniform. Therefore, they used a wide variety of assistive devices to achieve
independent mobility, including ankle–foot orthoses, walking sticks, four-wheeled
walkers and sometimes wheelchairs or mobility scooters. Device usage among these
groups also varied depending on environmental context e.g. an individual may have
used a walking stick indoors, a four-wheeled walker outdoors and a wheelchair for
long distances. Additionally, all PUs with stroke had upper limb impairments, which
influenced the range of assistive devices which were feasible for them to use.
All frail PUs used a four-wheeled walker, although some preferred to use
smaller devices like a walking stick around the home for convenience. Only one
pre-frail PU regularly used devices to assist with mobility, however, those who did
not use devices sought assistance from other persons for specific mobility tasks e.g.
getting on and off buses. All PUs, except frail older adults, expressed concerns
about their balance and risk of falling. While this finding may seem counter-
intuitive, it can be attributed to a lack of risk exposure among frail PUs. Only one
frail PU was independently mobile in the community – all others either required
supervision or did not go out.
Inside the home, almost all PUs reported difficulties with housekeeping acti-
vities which involved standing while performing tasks with their hands e.g. food
preparation and cooking. Many PUs reported sitting to complete these tasks, since
their walking aids could not provide the necessary hands-free support. PUs across all
groups reported receiving assistance from family or caregivers for housework such
as cleaning, vacuuming and laundry. A number of PUs also reported having home
adaptations to facilitate independent mobility e.g. ramps in/out of the home, stair
lifts, grab rails and adapted toilet and shower seats. Gardening was also challenging
for many PUs – some due to problems negotiating uneven terrain, and others due to
issues with bending or crouching down, and rising from such positions.
In the community, all PUs experienced limitations in walking long distances,
walking in crowded environments, negotiating stairs, and performing any activities
that required hand/arm function while using a walking aid e.g. shopping. Most PUs
relied on others for transport, as they did not or could not drive independently.
SUs had previous experience with a number of mobility aids, particular
walking sticks, crutches, four-wheeled walkers and wheelchairs, as well as ankle–
foot orthoses among SUs who worked with stroke populations. Most SUs reported
positive experiences, although challenges reported included the limited contexts in
which some devices can be used, lack of hands-free mobility aids, the bulkiness of
some devices, difficulty in training PUs in the correct usage of a device, and limited
access to devices due to costs or reimbursement options.
 Exploring user requirements 83

4.5.2.2 Design requirements for wearable assistive devices

for mobility
Table 4.4 summarises the main design requirements noted by PUs and SUs for a
wearable device to assist mobility. Design features are prioritised as ‘must-have’
and ‘nice-to-have’ based on the frequency and emphasis with which they were cited

Table 4.4 Prioritised design requirements for potential users of a soft lower body
assistive exoskeleton

Primary users Secondary users

Must-have design features
Effective Improves safety and stability for users,
preventing falls during gait and transfers
Facilitates better quality, safer, faster Improves users’ independence
walking, for longer and with less effort
Safe to use in the presence of other devices Usable in a range of different environ-
and health conditions ments (home, clinic, community, etc.)
Hands-free usage Adaptable to different users and their
individual needs for assistance
Supports the appropriate joints/areas of Easy to use (particularly for PUs with
the lower limb, as required by the user cognitive impairment)
Help with foot drop and coordination Easy to don and doff (even for users with
during gait (PUs with stroke) upper limb impairments)
Easy to don and doff Adjustable sizing
Worn under users’ usual clothing No extensive training and support required
to use
Comfortable User support available to PUs and SUs,
if required
Lightweight Lightweight
Easy to clean Robust
Should not limit the user’s own move- Compatible with usual footwear and
ments clothing
Easy to use Easy to clean
Compatible with users’ preferred clothing Low cost/reimbursed through public and/
or private schemes
Not bulky to wear
Low cost, or accessible at no cost to PU
Nice-to-have design features
Provides help and support with bending Improves users’ confidence in mobility
activities and transfers
Robust materials PU can don and doff independently
Not bulky to store Worn under users’ clothing
Possible to use independently Not bulky, particularly if used in the home
84 Wearable exoskeleton systems: design, control and applications

by users of each type i.e. features noted by multiple users and/or highlighted as
being vitally important were considered as ‘must-haves’; features noted by very
few users or considered as being not entirely necessary were considered to be ‘nice-
to-have’. Both PUs and SUs cited very similar requirements overall, with some
notable discrepancies.
The effectiveness of the device in improving mobility and the safety of the
device were high priorities for both user groups. Effectiveness was favoured by all
PUs, whereas SUs felt that improving user safety – including prevention of falls –
was the most important function of a device to assist mobility.
Both PUs and SUs felt strongly that being easy to use was an essential feature
of any mobility-related assistive device. This includes being sufficiently simple and
intuitive to use for PUs with cognitive impairments, and being easy to handle for
users with upper limb impairments.
Of particular relevance to wearable devices was the requirement to be easy to
don and doff, as noted by both PUs and SUs. Both groups felt that it should ideally
be possible to don and doff a device independently, although PUs with stroke/SCI
felt that requiring some assistance from another person would be acceptable,
likely due to the higher prevalence and severity of upper limb impairments among
these users.
Mixed opinions on aesthetics were noted; some PUs felt that appearance was
very important, while others felt that a device does not have to look nice, as long as
it is effective. SUs considered aesthetics to be more important if a device is to be
used in community settings, rather than in hospital or rehabilitation settings, due to
the potential for stigma to be felt by PUs in the community. Despite these mixed
opinions, both user groups agreed that a device should be lightweight and not
excessively bulky. Also, PUs with stroke/SCI noted that a device should be com-
patible with their usual clothing, as many had experienced problems using ankle–
foot orthoses with their usual footwear and trousers.
Both user groups agreed that a device must be affordable, either by being available
to purchase at low cost, or being available at no cost to PUs via public health services or
reimbursement by health insurance companies. For expensive devices, SUs believed
that reimbursement was critical to facilitate uptake, as they believed that users would
very rarely have the means or willingness to pay for such devices.

4.5.2.3 User perspectives on a soft assistive exoskeleton concept

Most PUs had positive opinions on the XoSoft concept, and felt that it could be
helpful for assisting walking and general mobility. The hands-free nature of the
device was seen as a distinct benefit over current mobility aids. PUs with specific
needs perceived other more specific benefits e.g. PUs with stroke felt it could help
with foot drop and lower limb coordination during gait.
Frail older adult PUs were the least positive about the concept, with only one such
individual expressing an interest in using such a device. They perceived few benefits to
the concept over and above their current mobility aids. Most frail older PUs felt that
XoSoft would be too complex for them to use, too difficult to don and doff, and
considered learning to use such a device to be an unnecessary burden for them.
 Exploring user requirements 85

In general, SUs perceived a large range of positive functions and benefits to

the XoSoft concept, particularly the flexibility and the adaptive nature, which
could allow it to be used by many different patients with varying conditions in a
range of settings.
SUs working in neurological rehabilitation had mixed opinions on the use of
XoSoft, as some felt there was a sufficient array of exoskeleton products available
in this field already. However, a device which could be used outside of a laboratory
setting, takes little time to set up for use and can provide adjustable partial support
to users was perceived as being beneficial, as it would offer functions beyond those
of existing products.
The main concerns of SUs with XoSoft were ease of donning, doffing and
general ease of use, particularly in frail users or those with cognitive impairments.
SUs acknowledged the technical challenges of creating a sufficiently simple, yet
reliable and effective interface for both PUs and SUs.
SUs also recognised the potential to use a soft lower body assistive exoskeleton
with individuals outside the specific PU groups targeted in the XoSoft project.
These broader potential applications could include use as a rehabilitation tool rather
than as an assistive device for multiple patient populations in clinics and hospitals
and for home and/or clinical use in musculoskeletal rehabilitation.

4.6 User needs: implications for soft exoskeleton design

The study described in Section 4.5 has revealed information of value to all design
stakeholders on the requirements of potential PUs and SUs of a lower body soft exos-
keleton to assist adults with mobility impairments. The study generated unique insights
into users’ desired functional and design requirements for such devices, from which
many opportunities and challenges for soft exoskeleton technologies can be drawn.

4.6.1 Functional requirements

A clear understanding of users’ design requirements (e.g. Table 4.4) is necessary to
develop an effective device, since effectiveness is relative to the user’s intended
function or goal [57]. For PUs, an effective device would support and assist them in
obtaining gross improvements in overall mobility and carrying out activities of
daily living with greater ease and independence. SUs placed a greater emphasis on
improving user safety and preventing falls. These findings echo those of a previous
survey of wheelchair users’ and healthcare professionals’ perspectives on exoske-
leton technology [58]. The emphasis on fall prevention and risk minimisation from
healthcare professionals is understandable, given the proliferation of evidence,
clinical guidelines and policies on this topic in recent years [59–61]. It is clear that
providing adequate assistance to users while ensuring stability and safety are the
key functional requirements.
It is also imperative that wearable devices that aim to assist mobility incor-
porate design features that support – and certainly do not impede – users’ ability to
undertake other activities. For example, aside from walking, support for bending
86 Wearable exoskeleton systems: design, control and applications

and crouching activities were highlighted as priority functions by PUs in the pre-
sent study and in a previous study of older adults’ requirements of a rigid assistive
exoskeleton [62]. There are considerable design and technical challenges asso-
ciated with developing a user-friendly system that provides adequate assistance
without restricting intended movements, while maintaining dynamic stability and
ensuring user safety.

4.6.2 Design and aesthetics

A notable design feature sought after by the majority of users in the present study
was the possibility to wear a device under their usual clothing. This is not a com-
mon feature of current exoskeletons, possibly because it would be unfeasible for
bulky, rigid exoskeletons. However, when designing a device for use in home and
community environments, aesthetics cannot be dismissed as a trivial aspect of
device design. In home and community contexts, an assistive device may be a
source of stigma for users [57], more so than in clinical contexts, a matter high-
lighted by SUs in the present study. This presents a distinct opportunity for light-
weight, soft exoskeletons to be developed for use in the home and the community.
It also emphasises the need to continually seek to reduce the size, weight and noise
of rigid exoskeletons to maximise their viability in these contexts.
As seen previously in relation to rigid exoskeleton design, comfort and ease of
putting on/taking off the device were of great concern to all users [58]. Ease of
donning and doffing the device was greatly emphasised in the present study, most
likely due to the high prevalence of upper limb impairments among the PUs
interviewed and SUs’ extensive experience working with such individuals. Soft
exoskeleton technologies offer great potential to meet these users’ needs, as they
present opportunities to integrate sensing and actuation into comfortable light-
weight garments that are fit for everyday life, can be easily donned and doffed, and/
or worn for moderate periods of time. Further research to identify optimal soft
exoskeleton garment designs to facilitate users with a variety of impairments is
required.

4.6.3 Willingness to use the concept

The majority of PUs and SUs in the present study were positive about the concept
of a soft assistive exoskeleton and indicated willingness to use such a device pro-
vided it meets their functional and design requirements. Frail older PUs were the
least positive about the prospect of using such a device. These individuals were
considerably older than other PU groups, had a greater number of co-morbidities,
and experienced greater levels of disability and low levels of social participation.
These factors can all contribute to these users’ perceptions of the usability and use-
worthiness of an assistive device, particularly when considering the activities
which the device aims to assist [57]. For example, if a frail PU with poor general
health considers increasing their mobility to be neither possible to do, nor worth
doing, then using a device to assist mobility will not be a priority.
 Exploring user requirements 87

Cost was identified as a concern for both PUs and SUs, in line with previous
user studies on assistive exoskeletons [58] and can also affect willingness to use a
concept. While soft exoskeletons are not yet at a sufficient technology readiness
level to be placed on the market, potential avenues to create accessible products
should be considered as part of overall PSS development. For example, offering
leasing and or rental services to lower costs of ownership for PUs, while incor-
porating refurbishment and circular economy options for TUs.

4.6.4 Alternative assistive devices

Any proposed exoskeleton intended for use by stroke and SCI populations will face
competition from existing exoskeleton products aimed at these user groups.
Therefore, designers of new exoskeletons in this space must strive to create systems
which offer new and/or improved features in comparison to existing products. The
limited range of currently available devices are rigid exoskeletons for use in the
rehabilitation of individuals with moderate–severe motor impairments and pre-
dominantly provide full assistance and support to users, while generally being only
suitable for use in an intensive therapeutic context [63,64]. Opportunities therefore
exist for assistive devices, devices for users with low–moderate levels of impair-
ment, devices that offer adaptable support and devices which can be used in home
or community settings. Also, considering the comfort and wearability requirements
of users (Table 4.4), current hard exoskeletons are unlikely to be readily accepted
by PUs with low–moderate levels of impairment for use in activities of daily living.
A soft exoskeleton which provides adaptable support and can be used in the home
and/or community would therefore provide a unique and innovative solution for
users, with greater probability of user acceptance.

4.6.5 Current challenges for soft exoskeleton technologies

Soft robotics is an emerging area with a wide variety of potential applications
across many fields, however, the technologies are very much still in development
[65,66]. At present, it is difficult to foresee where the limits of soft robotic tech-
nologies lie, and thus what solutions may be realised or may not be possible, as the
field is rapidly advancing. Considering user requirements together with technolo-
gical limitations, some of the challenges that currently exist with respect to soft
exoskeletons as assistive devices include:
● Soft sensing (motion): The use of soft sensors to measure the motion of the
human body is relatively easier than their use in the measurement of soft
robots: the human body can be considered as a relatively rigid structure in
comparison to soft robots, which present highly complex motions with infinite
degrees of freedom and non-linearities. Even where human joint motion is not
ideal, it can often be approximated by a hinge or other simple kinematic chain.
The challenge of measuring human joint motion using soft sensors lies in the
fact that we need to measure it through indirect variables such as elongation
and bending [67], since it is not always possible to place a sensor in the centre
88 Wearable exoskeleton systems: design, control and applications

of rotation of the human joint. For this reason, and in order to get representa-
tive information, a detailed calibration process is required to correlate the
sensor reading with the joint angle. This calibration might also include the
identification of hysteresis and other non-linear behaviours.
Most technologies are based on the electrical properties of certain materials
which change under stress or strain conditions. These types of sensors can be
classified as resistive and capacitive sensors. However, it is possible to use the
measurement of other physical variables such as light transmission e.g. using
optic fibres [68].
Another common challenge of measuring motion using soft sensors is the
placement of the sensors, since it is not always possible to know the exact
location of the sensor over time. For this reason, sensory fusion algorithms are
required to combine the information coming from a redundant set of sensors to
get a sufficiently reliable estimation of the motion.
● Soft sensing (force/pressure): Similar technologies as those used to
measure motion can be used to measure force/pressure in a soft lower body
exoskeleton, since the body areas where force/pressure will be measured
will – in most cases – present very low levels of deformation [69]. Typi-
cally, the measurement of the external forces is only required on the
contact between the foot and the floor. The level of information required
depends on the design of the controller. A simple sensor that is able to
detect the heel-strike and the toe-off moments is the minimum required for
reliable gait detection. The robust measurement of the contact force at the
feet is still a technical challenge, even more when considering the detec-
tion of the direction of the force.
● Soft actuation: Pneumatic systems have been widely used for soft actuation in
wearable robots. Artificial muscles – comprised of pneumatic components
combined with structured materials – have been successfully used for the
actuation of assistive exoskeletons [70]. Their power–weight ratio is excellent
(500–2,000 W/kg), and they can transmit high forces (100–500 N/cm2) at a
reasonable level of efficiency (32%–50%) [71]. The behaviour and interaction
is quite natural; however, the control bandwidth is limited (typically around
5 Hz), due to limitations of the flow rate [72]. These systems also require of a
source of pressure, commonly an air compressor or pressurized gas tank, which
is not easy to integrate on a wearable device, and the level of noise is not
suitable for some applications e.g. for home or community use.
Cable-driven actuation is a common alternative which permits the use of
classical actuation systems (electrical motors) in a soft wearable device.
The actuators are placed in a rigid structure, usually carried on the back of
the user, while the motion/force is transmitted to the target joints by means
of cables [73]. These systems can provide a high level of assistance, only
limited by the force exchange between the wearable device and the user.
The weight and energy efficiency of cable-driven actuation depends mostly on
the electrical motors and mechanical components used in the system, which
can be in the order of 100–300 N m/kg for commercial motors.
 Exploring user requirements 89

Other alternatives for soft actuation systems in wearable devices exist, but this
is still an open research topic with scarce applications in real scenarios.
Technologies such as Shape Memory Alloys or Electroactive Polymers look
promising, but it remains challenging to implement them on wearable devices
[74]. However, when a low–moderate level of assistance is required, it is
possible to use quasi-passive elements such as variable damping [75], variable
stiffness, clutches or elastic materials, which can be implemented using non-
classical actuation systems. In practice, the challenge of providing sufficient
support for specific users and their varied individual needs over time will be
highly complex (e.g. in stroke), since the requirements for support will differ
between users depending on the nature and extent of their pathology.
● Force transmission: The assistive forces generated by the wearable device
will be transmitted to the user through the body attachments. One of the major
challenges is that the human-body surface is highly heterogeneous; each
individual user is completely different, even more so when considering users
with various pathologies. This makes the design of the body attachments
highly important and challenging, since their design will limit the total amount
of force the system can provide to the user. In most of the cases, the body
attachments are designed taking advantage of designs present in other areas,
such as orthosis, wheelchair postural supports and trekking backpacks.
It is also important to consider that all the forces generated by the system
will load the human rigid structure (bones and joints) because of the absence of
a rigid structure in the wearable device. For this reason, the maximum
expected assistive force cannot exceed the one generated by the human
muscular system. That means, using a soft wearable assistive device, in the
best case scenario, can restore the mobility of a disabled limb, but we cannot
augment the capabilities of the human body.
● Control: The same kind of algorithms used in hard exoskeletons can be
applied to soft wearable devices, but with some added challenges. As indicated
previously, the estimation of the motion and interaction forces might not be
fully reliable. Also, due to the soft nature of the system, the actuation may have
a longer reaction time. However, the control system should maintain a high
level of stability to avoid any discomfort for the user.
● Human-robot interaction: The use of soft robotics in assistive exoskeletons
may offer improved safety over rigid exoskeletons when interacting with
humans, but the aforementioned issues with sensing, actuation and control will
also create more complexity when considering the safety of human-robot
interaction.
Current technologies allow soft exoskeletons to provide low–moderate assistance,
which would be appropriate for assisting users with low–moderate levels of
impairment in most common tasks (e.g. standing, walking, etc.). These levels of
assistance can represent a partial restoration of the gait pattern, implying an
improvement of the stability and reduction of the energy expenditure, which is,
in most of the cases, a major improvement in the quality of life of the patient.
90 Wearable exoskeleton systems: design, control and applications

The short-term future of soft assistive exoskeletons must involve compromise

between the high-level demands and expectations of users and the technical feasi-
bility of realising these demands [76]. The long-term future looks promising,
considering the user requirements stated above. With further advances in soft
assistive technologies more assistance could be provided for more complex tasks
(e.g. sit-to-stand, stair climbing, etc.), with the ultimate aim of creating safe to use
and user-friendly product-service systems.

4.7 Chapter summary

Using the framework of a project aiming to develop a soft lower body exoskeleton
to assist mobility, this chapter has presented:
● An overview of the main principles of user-centred design;
● An example of the use of a mixed methods study design to engage users in soft
exoskeleton design by exploring their needs and design requirements;
● Insights into the perspectives of users with mild–moderate mobility impair-
ments, and individuals close to these users, on their requirements of an
assistive device for mobility, and the implications of these for exoskeleton
technologies;
● The main opportunities and challenges currently facing soft exoskeletons as
assistive devices.

References
[1] Davis F.D., Bagozzi R.P., Warshaw P.R., ‘User acceptance of computer
technology: a comparison of two theoretical models’. Management Science,
1989. 35(8): pp. 982–1003.
[2] Davis Jr F.D., A technology acceptance model for empirically testing new
end-user information systems: Theory and results. 1986, Massachusetts
Institute of Technology, Cambridge, MA.
[3] Venkatesh V., Davis F.D., ‘A theoretical extension of the technology
acceptance model: Four longitudinal field studies’. Management Science,
2000. 46(2): pp. 186–204.
[4] Cook A.M., Polgar J.M., Assistive technologies – E-Book: Principles and
practice. 4th ed., 2014, Elsevier Health Sciences.
[5] Shah S.G.S., Robinson I., AlShawi S., ‘Developing medical device tech-
nologies from users’ perspectives: a theoretical framework for involving
users in the development process’. International Journal of Technology
Assessment in Health Care, 2009. 25(04): pp. 514–521.
[6] de Eyto A., Ryan A., McMahon M., Hassett G., Flynn M., ‘Health futures
lab: transdisciplinary development of T shaped professionals through
‘‘wicked problem’’ challenges’. 8th Engineering Education for Sustainable
Development (EESD), UBC, Vancouver, Canada, 2015.
 Exploring user requirements 91

[7] Abras C., Maloney-Krichmar D., Preece J., ‘User-centered design’. Bain-
bridge, W. Encyclopedia of Human-Computer Interaction. Thousand Oaks:
Sage Publications, 2004. 37(4): pp. 445–456.
[8] Eurostat, European Health and Social Integration Survey. 2012.
[9] Bateni H., Maki B.E., ‘Assistive devices for balance and mobility: benefits,
demands, and adverse consequences’. Archives of Physical Medicine and
Rehabilitation, 2005. 86(1): pp. 134–145.
[10] 1Bradley S.M., Hernandez C.R., ‘Geriatric assistive devices’. American
Family Physician, 2011. 84(4): pp. 405–411.
[11] Hoenig H., Morgan M., Montgomery C., Landerman L.R., Caves K., ‘One
size does not fit all-mobility device type affects speed, collisions, fatigue,
and pain’. Archives of Physical Medicine and Rehabilitation, 2015. 96(3):
pp. 489–497.
[12] Stevens J.A., Thomas K., Teh L., Greenspan A.I., ‘Unintentional fall injuries
associated with walkers and canes in older adults treated in US emergency
departments’. Journal of the American Geriatrics Society, 2009. 57(8):
pp. 1464–1469.
[13] Feigin V.L., Forouzanfar M.H., Krishnamurthi R., et al., ‘Global and
regional burden of stroke during 1990–2010: findings from the Global
Burden of Disease Study 2010’. The Lancet, 2014. 383(9913): pp. 245–255.
[14] Lawrence E.S., Coshall C., Dundas R., et al., ‘Estimates of the prevalence
of acute stroke impairments and disability in a multiethnic population’.
Stroke, 2001. 32(6): pp. 1279–1284.
[15] Woolley S.M., ‘Characteristics of gait in hemiplegia’. Topics in Stroke
Rehabilitation, 2001. 7(4): pp. 1–18.
[16] Lamontagne A., Malouin F., Richards C.L., Dumas F., ‘Mechanisms of
disturbed motor control in ankle weakness during gait after stroke’. Gait &
Posture, 2002. 15(3): pp. 244–255.
[17] Roche N., Bonnyaud C., Geiger M., Bussel B., Bensmail D., ‘Relationship
between hip flexion and ankle dorsiflexion during swing phase in chronic
stroke patients’. Clinical Biomechanics (Bristol, Avon), 2015. 30(3):
pp. 219–225.
[18] Burdett R.G., Borello-France D., Blatchly C., Potter C., ‘Gait comparison of
subjects with hemiplegia walking unbraced, with ankle–foot orthosis, and
with Air-Stirrup brace’. Physical Therapy, 1988. 68(8): pp. 1197–1203.
[19] Yamamoto S., Ibayashi S., Fuchi M., Yasui T., ‘Immediate-term effects of
use of an ankle–foot orthosis with an oil damper on the gait of stroke patients
when walking without the device’. Prosthetics and Orthotics International,
2015. 39(2): pp. 140–149.
[20] Chen G., Patten C., Kothari D.H., Zajac F.E., ‘Gait differences between
individuals with post-stroke hemiparesis and non-disabled controls at mat-
ched speeds’. Gait Posture, 2005. 22(1): pp. 51–56.
[21] Stanhope V.A., Knarr B.A., Reisman D.S., Higginson J.S., ‘Frontal plane
compensatory strategies associated with self-selected walking speed in
92 Wearable exoskeleton systems: design, control and applications

individuals post-stroke’. Clinical Biomechanics (Bristol, Avon), 2014. 29(5):

pp. 518–522.
[22] Cruz T.H., Dhaher Y.Y., ‘Impact of ankle–foot-orthosis on frontal plane
behaviors post-stroke’. Gait & Posture, 2009. 30(3): pp. 312–316.
[23] Kim C.M., Eng J.J., ‘Magnitude and pattern of 3D kinematic and kinetic gait
profiles in persons with stroke: relationship to walking speed’. Gait & Pos-
ture, 2004. 20(2): pp. 140–146.
[24] Connell L.A., Lincoln N., Radford K., ‘Somatosensory impairment after
stroke: frequency of different deficits and their recovery’. Clinical Rehabi-
litation, 2008. 22(8): pp. 758–767.
[25] Rowe F., Brand D., Jackson C.A., et al., ‘Visual impairment following
stroke: do stroke patients require vision assessment?’. Age and Ageing, 2009.
38(2): pp. 188–193.
[26] Lord S.E., McPherson K., McNaughton H.K., Rochester L., Weatherall M.,
‘Community ambulation after stroke: how important and obtainable is it and
what measures appear predictive?’. Archives of Physical Medicine and
Rehabilitation, 2004. 85(2): pp. 234–239.
[27] Patel M.D., Coshall C., Rudd A.G., Wolfe C.D., ‘Cognitive impairment
after stroke: clinical determinants and its associations with long-term stroke
outcomes’. Journal of the American Geriatrics Society, 2002. 50(4):
pp. 700–706.
[28] Tatemichi T., Desmond D., Stern Y., Paik M., Sano M., Bagiella E., ‘Cog-
nitive impairment after stroke: frequency, patterns, and relationship to
functional abilities’. Journal of Neurology, Neurosurgery & Psychiatry,
1994. 57(2): pp. 202–207.
[29] Furlan J.C., Sakakibara B.M., Miller W.C., Krassioukov A.V., ‘Global
incidence and prevalence of traumatic spinal cord injury’. Canadian Journal
of Neurological Sciences, 2013. 40(4): pp. 456–464.
[30] National Spinal Cord Injury Statistical Center, Spinal Cord Injury (SCI)
Facts and Figures at a Glance. 2016; Available from: https://www.nscisc.
uab.edu/Public/Facts%202016.pdf.
[31] Van Hedel H., Wirz M., Dietz V., ‘Standardized assessment of walking
capacity after spinal cord injury: the European network approach’. Neuro-
logical Research, 2008. 30(1): pp. 61–73.
[32] Louie D.R., Eng J.J., Lam T., Spinal Cord Injury Research Evidence
Research Team, ‘Gait speed using powered robotic exoskeletons after
spinal cord injury: a systematic review and correlational study’. Journal of
NeuroEngineering and Rehabilitation, 2015. 12: pp. 82.
[33] Fisahn C., Aach M., Jansen O., et al., ‘The effectiveness and safety of
exoskeletons as assistive and rehabilitation devices in the treatment of neu-
rologic gait disorders in patients with spinal cord injury: a systematic
review’. Global Spine Journal, 2016. 6(8): pp. 822–841.
[34] Grasmucke D., Zieriacks A., Jansen O., et al., ‘Against the odds: what to
expect in rehabilitation of chronic spinal cord injury with a neurologically
controlled Hybrid Assistive Limb exoskeleton. A subgroup analysis of
 Exploring user requirements 93

55 patients according to age and lesion level’. Neurosurgical focus, 2017.

42(5): pp. E15.
[35] New P., Simmonds F., Stevermuer T., ‘A population-based study comparing
traumatic spinal cord injury and non-traumatic spinal cord injury using a
national rehabilitation database’. Spinal Cord, 2011. 49(3): pp. 397–403.
[36] New P.W., Rawicki H.B., Bailey M.J., ‘Nontraumatic spinal cord injury:
demographic characteristics and complications’. Archives of Physical Med-
icine and Rehabilitation, 2002. 83(7): pp. 996–1001.
[37] United Nations Department of Economic and Social Affairs Population
Division, World Population Ageing 2013, New York: United Nations. 2013.
[38] Salive M.E., ‘Multimorbidity in older adults’. Epidemiologic Reviews, 2013.
35: pp. 75–83.
[39] Gijsen R., Hoeymans N., Schellevis F.G., Ruwaard D., Satariano W.A., van
den Bos G.A., ‘Causes and consequences of comorbidity: a review’. Journal
of Clinical Epidemiology, 2001. 54(7): pp. 661–674.
[40] Hung W.W., Ross J.S., Boockvar K.S., Siu A.L., ‘Recent trends in chronic
disease, impairment and disability among older adults in the United States’.
BMC Geriatrics, 2011. 11: pp. 47.
[41] Chatterji S., Byles J., Cutler D., Seeman T., Verdes E., ‘Health, functioning,
and disability in older adults – present status and future implications’.
The Lancet, 2015. 385(9967): pp. 563–575.
[42] Xue Q.L., ‘The frailty syndrome: definition and natural history’. Clinics in
Geriatric Medicine, 2011. 27(1): pp. 1–15.
[43] Fried L.P., Tangen C.M., Walston J., et al., Cardiovascular Health Study
Collaborative Research Group, ‘Frailty in older adults: evidence for a phe-
notype’. The Journals of Gerontology. Series A, Biological Sciences and
Medical Sciences, 2001. 56(3): pp. M146–M156.
[44] Ferri C.P., Prince M., Brayne C., et al., ‘Global prevalence of dementia:
a Delphi consensus study’. The Lancet, 2005. 366(9503): pp. 2112–2117.
[45] Busse A., Bischkopf J., Riedel-Heller S.G., Angermeyer M.C., ‘Mild
cognitive impairment: prevalence and incidence according to different
diagnostic criteria. Results of the Leipzig Longitudinal Study of the Aged
(LEILA75þ)’. British Journal of Psychiatry, 2003. 182: pp. 449–454.
[46] Storz-Pfennig P., Schmedders M., Dettloff M., ‘Trials are needed before new
devices are used in routine practice in Europe’. BMJ: British Medical
Journal, 2013. 346: f1646.
[47] Faruqui S.R., Jaeblon T., ‘Ambulatory assistive devices in orthopaedics:
uses and modifications’. Journal of the American Academy of Orthopaedic
Surgeons, 2010. 18(1): pp. 41–50.
[48] de Vries O.J., Peeters G.M., Elders P.J., et al., ‘Multifactorial intervention
to reduce falls in older people at high risk of recurrent falls: a
randomized controlled trial’. Archives of Internal Medicine, 2010. 170(13):
pp. 1110–1117.
[49] Wilhelmson K., Andersson C., Waern M., Allbeck P., ‘Elderly people’s per-
spectives on quality of life’. Ageing & Society, 2005. 25(04): pp. 585–600.
94 Wearable exoskeleton systems: design, control and applications

[50] Pettersson I., Berndtsson I., Appelros P., Ahlström G., ‘Lifeworld perspec-
tives on assistive devices: Lived experiences of spouses of persons with
stroke’. Scandinavian Journal of Occupational Therapy, 2005. 12(4):
pp. 159–169.
[51] Chen T.-Y.A., Mann W.C., Tomita M., Nochajski S., ‘Caregiver involve-
ment in the use of assistive devices by frail older persons’. OTJR: Occupa-
tion, Participation and Health, 2000. 20(3): pp. 179–199.
[52] Sun J.-H., Tan L., Yu J.-T., ‘Post-stroke cognitive impairment: epidemiol-
ogy, mechanisms and management’. Annals of Translational Medicine,
2014. 2(8): p. 80.
[53] Folstein M.F., Folstein S.E., McHugh P.R., ‘‘‘Mini-mental state’’. A prac-
tical method for grading the cognitive state of patients for the clinician’.
Journal of Psychiatric Research, 1975. 12(3): pp. 189–198.
[54] Holden M.K., Gill K.M., Magliozzi M.R., ‘Gait assessment for neurologi-
cally impaired patients. Standards for outcome assessment’. Physical Ther-
apy, 1986. 66(10): pp. 1530–1539.
[55] Middleton A., Fritz S.L., Lusardi M., ‘Walking speed: the functional vital
sign’. Journal of Aging and Physical Activity, 2015. 23(2): pp. 314–322.
[56] Mahoney F.I., Barthel D., ‘Functional evaluation: the Barthel Index’.
Maryland State Medical Journal, 1965. 14: pp. 56–61.
[57] Krantz O., ‘Assistive devices utilisation in activities of everyday life – a
proposed framework of understanding a user perspective’. Disability and
Rehabilitation: Assistive Technology, 2012. 7(3): pp. 189–198.
[58] Wolff J., Parker C., Borisoff J., Mortenson W., Mattie J., ‘A survey of sta-
keholder perspectives on exoskeleton technology’. Journal of NeuroEngi-
neering and Rehabilitation, 2014. 11(1): pp. 169.
[59] American Geriatrics Society, British Geriatrics Society, ‘Summary of the
Updated American Geriatrics Society/British Geriatrics Society Clinical
Practice Guideline for Prevention of Falls in Older Persons’. Journal of the
American Geriatrics Society, 2011. 59(1): pp. 148–157.
[60] Avin K.G., Hanke T.A., Kirk-Sanchez N., et al., Academy of Geriatric
Physical Therapy of the American Physical Therapy Association, ‘Man-
agement of falls in community-dwelling older adults: clinical guidance
statement from the Academy of Geriatric Physical Therapy of the American
Physical Therapy Association’. Physical Therapy, 2015. 95(6): pp. 815–834.
[61] NICE, Falls: assessment and prevention of falls in older people. NICE
clinical guideline 161. 2013, National Institute for Health and Care Excel-
lence, Manchester.
[62] O’Sullivan L., Power V., Virk G., et al., ‘End user needs elicitation for a
full-body exoskeleton to assist the elderly’. Procedia Manufacturing, 2015. 3:
pp. 1403–1409.
[63] Sale P., Franceschini M., Waldner A., Hesse S., ‘Use of the robot assisted
gait therapy in rehabilitation of patients with stroke and spinal cord injury’.
European Journal of Physical and Rehabilitation Medicine, 2012. 48(1):
pp. 111–121.
 Exploring user requirements 95

[64] Chen G., Chan C.K., Guo Z., Yu H., ‘A review of lower extremity assistive
robotic exoskeletons in rehabilitation therapy’. Critical Reviews in Bio-
medical Engineering, 2013. 41(4–5): pp. 343–363.
[65] Trivedi D., Rahn C.D., Kier W.M., Walker I.D., ‘Soft robotics: biological
inspiration, state of the art, and future research’. Applied Bionics and Bio-
mechanics, 2008. 5(3): pp. 99–117.
[66] Majidi C., ‘Soft robotics: a perspective—current trends and prospects for the
future’. Soft Robotics, 2014. 1(1): pp. 5–11.
[67] Lucarotti C., Totaro M., Sadeghi A., Mazzolai B., Beccai L., ‘Revealing
bending and force in a soft body through a plant root inspired approach’.
Scientific Reports, 2015. 5: pp. 8788.
[68] Salceda-Delgado G., Van Newkirk A., Antonio-Lopez J., Martinez-Rios A.,
Schülzgen A., Correa R.A., ‘Compact fiber-optic curvature sensor based on
super-mode interference in a seven-core fiber’. Optics Letters, 2015. 40(7):
pp. 1468–1471.
[69] Viry L., Levi A., Totaro M., et al., ‘Flexible three-axial force sensor for
soft and highly sensitive artificial touch’. Advanced Materials, 2014. 26(17):
pp. 2659–2664.
[70] Tsagarakis N.G., Caldwell D.G., ‘Development and control of a ‘‘soft-
actuated’’ exoskeleton for use in physiotherapy and training’. Autonomous
Robots, 2003. 15(1): pp. 21–33.
[71] Tsagarakis N., Caldwell D.G. Improved modelling and assessment of pneu-
matic muscle actuators. Robotics and Automation, 2000. Proceedings.
ICRA’00. IEEE International Conference on, 2000. IEEE.
[72] Davis S., Tsagarakis N., Canderle J., Caldwell D.G., ‘Enhanced modelling
and performance in braided pneumatic muscle actuators’. The International
Journal of Robotics Research, 2003. 22(3–4): pp. 213–227.
[73] Asbeck A.T., Dyer R.J., Larusson A.F., Walsh C.J., ‘Biologically-inspired
soft exosuit’. 2013 IEEE international conference on rehabilitation robotics
(ICORR), 2013, IEEE. pp. 1–8.
[74] Cianchetti M., Mattoli V., Mazzolai B., Laschi C., Dario P., ‘A new design
methodology of electrostrictive actuators for bio-inspired robotics’. Sensors
and Actuators B: Chemical, 2009. 142(1): pp. 288–297.
[75] Hauser S., Robertson M.C., Ijspeert A., Paik J. ‘JammJoint: A variable
stiffness device based on granular jamming for wearable joint support’.
IEEE Robotics and Automation Letters, 2017. DOI: 10.1109/LRA.2017.
2655109.
[76] Viteckova S., Kutilek P., Jirina M., ‘Wearable lower limb robotics:
A review’. Biocybernetics and Biomedical Engineering, 2013. 33(2):
pp. 96–105.