Exoskeletons as an Assistive Technology for Mobility and Manipulation
Jaimie Borisoff, Mahsa Khalili, W. Ben Mortenson, and H. F. Machiel Van der Loos
Critical Review of the Available Technology 185
Lower Limb Exoskeletons 190
Upper Limb Exoskeletons 195
Critical Review of the Available Utilization Protocols: How Can
Exoskeletons Be Used? 199
Lower Limb Exoskeletons 200
Upper Limb Exoskeletons 202
Review of User Studies, Outcomes, and Clinical Evidence 203
Lower Limb Exoskeletons 203
Upper Limb Exoskeletons 208
Future Directions 209
After completing this chapter, readers will be able to
- 1. Describe the principles that underlie the performance criteria for a particular exoskeleton technology when it is implemented as an assistive technology for mobility or manipulation.
- 2. List the walking exoskeletons (lower limb exoskeletons, LLEs) that are either commercially available or nearing production.
- 3. List the upper limb exoskeletons (ULEs) for manipulation that are either commercially available or nearing production.
- 4. Describe and list the client populations, and their associated disabilities, that are candidate users for exoskeleton technologies.
- 5. Discuss the current limitations of exoskeletons for use in the community.
- 6. Describe the activities, applications, and environments suitable for exoskeleton use as a mobility device, both currently and in the future.
In 2013, it was estimated that over 41 million people in the United States (approximately 13% of the population) had a mobility disability, the most prevalent type of disability (Courtney-Long et al. 2015). In 2009, it was estimated that over 24 million people in the United States used assistive technology (AT) (Ilunga Tshiswaka et al. 2016), and in 2010, over 3.6 million used a wheelchair (Brault 2010). Wheelchairs, in particular, have progressed to the point that, regardless of their notoriety as a limiting factor, they are, in fact, empowering and paramount to many people in terms of achieving a high quality of life, independence, and participation in the community. Wheelchairs do present some problems when used chronically, however. Two problems are: (1) health (conventional, seated wheelchair use is associated with a variety of potential health concerns, such as skin breakdown and overuse injuries); and (2) access (the built and natural environments in which we live invariably impose barriers to full access and participation by wheelchair users). In addition, wheelchairs are simply not transformative; that is, they do not allow persons with a disability a level of mobility performance that approaches that of their nondisabled peers (Cowan et al. 2012).
As well as gait or mobility disabilities, some conditions affect manipulation. Manipulation, in this context, involves the use of arms, hands, and fingers to manipulate objects in the environment for a variety of reasons, but usually related to performing activities of daily living (ADLs), such as eating and interacting with computers, tasks that require a minimum level of upper limb functionality, including reaching and grasping. Upper extremity disorders can significantly reduce the independence of individuals and consequently decrease their quality of life. Some upper limb impairments can be reversed after undergoing intensive rehabilitation, whereas other disorders are progressive and may leave individuals with permanent impairment. Some of the latter disorders include brachial plexus injury (BPI), high-level spinal cord injury (SCI), stroke, multiple sclerosis (MS), amyotrophic lateral sclerosis (ALS), among other upper limb impairments affecting disabled and elderly populations. Just considering stroke, there are more than 6 million survivors in the United States alone, and stroke is the leading cause of serious long-term disability (Mozaffarian et al. 2016). It is well known that poor outcomes for upper limb function after stroke are common (Gowland 1982). Despite research that has been conducted in the area of upper limb functional improvement, there seems to be a substantial gap between commercially available solutions and the need in the community for viable devices.
To improve function for people with mobility and manipulation- related disabilities, exoskeletons have been developed and now represent an emerging AT sector. An exoskeleton, as defined in this chapter, is a powered robotic orthosis for people with disabilities. They are designed for mobility or manipulation and are wearable and autonomously operated; ideally, if used as AT, exoskeletons could be deployed in daily life scenarios (Van der Loos, Reinkensmeyer, and Guglielmelli 2015). More specifically, with regard to lower limb exoskeletons (LLEs), we use the definition of Louie et al. (2015, 1): “a multi-joint orthosis that uses an external power source to move at least two joints on each leg, which is portable, and can be used independent of a treadmill or body-weight support.” It is important to note that these are active devices; thus, we are not considering any passive exoskeletons or other types of orthoses.
While the focus of discussions on exoskeletons is usually on standing and walking, upper limb exoskeletons (ULEs) are being developed to improve manipulation and are also reviewed here. Manipulation is simply mobility of the upper extremities to perform a task, although, certainly with regard to medical applications, one significant distinguishing difference between ULE and LLE designs is that ULE designs are focused on precise control while LLEs are focused more on the power necessary for standing and walking. Consistent with the LLE definition cited, a definition for a ULE designed for manipulation would be that it is a wearable, portable, and autonomously operated multijoint orthosis that uses an external power source to move at least two joints on the upper limb for use in daily life scenarios.
In this chapter, we do not consider ULEs that need external attachment to a base of support (other than to a wheelchair), hence preserving their portability and assistive functionality. Exoskeletons with nonportable stable supports are considered to be robotic manipulators. Other robotic manipulators are portable (e.g., attached, like the JACO [Maheu et al. 2011], to a wheelchair). Robotic manipulators (Chapter 3) are meant to substitute movements for disabled upper extremities (Maciejasz et al. 2014), rather than assisting functionally natural movements of the limbs themselves. Robotic prostheses are addressed in Chapter 4. We also do not consider single-joint exoskeletons, such as ankle-only or knee-only powered orthoses (Lajeunesse et al. 2015).
Young and Ferris (2016) have categorized exoskeletons based on their intended use into three classes:
1. Augmentation of able-bodied people for increases in strength, speed, or endurance. This is usually associated with military applications, but industrial uses are proposed as well, including use by factory workers or health care professionals (e.g., for assisting patient transfers).
- 2. Rehabilitation or therapy devices, a class of exoskeletons that is outside the scope of this book.
- 3. Exoskeletons as an AT, as we consider in this chapter.
There is possible overlap in these three classes of technology; for example, a design with variable assistive control and seamless user-intentional control may be usable in the future by a person with a disability for rehabilitation and AT purposes, as well as used as an augmentation exoskeleton for military-industrial uses by an able-bodied person. There has been some effort in the literature to distinguish exoskeletons that were developed to expressly augment the physical capabilities of able-bodied individuals from powered orthoses. For instance, Herr has stated: “In general, the term ‘exoskeleton’ is used to describe a device that augments the performance of an able-bodied wearer, whereas the term ‘orthosis’ is typically used to describe a device that is used to assist a person with a limb pathology” (Herr 2009, 1). On the other hand, we propose to use the terminology adopted by the U.S. Food and Drug Administration (FDA), which recently approved the ReWalk exoskeleton, the first LLE to be approved, for supervised therapeutic use at home: “The device is assigned the generic name powered lower extremity exoskeleton, and it is identified as a prescription device that is composed of an external, powered, motorized orthosis that is placed over a person’s paralyzed or weakened limbs for medical purposes” (U.S. Food and Drug Administration, HHS 2015, 25227).
Almost exclusively, exoskeletons are being used today for rehabilitation purposes under supervision by a therapist. However, exoskeletons could actually become a replacement or adjunct for wheelchairs or other ATs. As of this writing (March 2016), there was little evidence to suggest this is likely in the short term. Nevertheless, this chapter discusses the potential for exoskeletons to become true mobility or manipulation assistive devices, reviews the products available today, and describes known efforts under way in research laboratories.
To develop a better understanding of the potential for exoskeletons to be used as mobility devices, it is important to understand how AT is defined. The Technology Related Assistance to Individuals with Disabilities Act of 1988 (U.S. Code Chapter 31 2011) described AT devices as “any item, piece of equipment, or product system, whether acquired commercially off the shelf, modified, or customized, that is used to increase, maintain, or improve functional capabilities of individuals with disabilities.” Similarly, the International Organization for Standardization (ISO) defined assistive products, in part, as any product (including devices, instruments, and equipment) used by persons with disability for participation, and to support or substitute for body functions, or to prevent activity limitations (ISO 2011). Although these definitions suggest that AT devices can be used for either compensation (environmentally focused interventions) or rehabilitation (interventions focused on changing/restoring the abilities of the individual), AT is usually conceived as a compensatory strategy (Russell et al. 1997). From this perspective, Cowan et al. (2012) described how some technologies, such as robotic exercise devices, may have an indirect effect on mobility by improving the capabilities of the user, whereas ATs have a direct effect on mobility without changing the body of the person. It is only when a device is used in the real world for purposeful activities that it becomes AT. That said, the relationship between compensation and rehabilitation is somewhat ambiguous, as devices such as manual wheelchairs may compensate for problems with ambulating, but at the same time provide a form of exercise, which may reduce the risk of more general deconditioning.
According to the Human Activity Assistive Technology (HAAT) model (Cook and Polgar 2015), successful AT prescription requires careful consideration of the person, the targeted activity, the AT that can be used, and the context in which it will be used. With the HAAT model, activity is a broad term that includes both the execution of a task and social participation as defined in the World Health Organization’s (WHO’s) International Classification of Functioning, Disability and Health (Svestkova 2008). For example, the provision of mobility AT requires a thorough understanding of the abilities of the individuals, activities that they want to perform (in terms of both mobility-related activities they want to take part in and activities they want to take part in using their mobility devices), the features of each device being considered, the accessibility of the environment, and availability of assistance.
The AT choices invariably involve compromises (Jutai et al. 2005). For example, someone may have difficulty self-propelling a manual wheelchair but might not want to use a power wheelchair because of the difficulty in transporting it in his or her current vehicle and would be unable to go places that did not have curb cuts (manual wheelchairs can be lifted up onto curbs). There may also be a temporal component to these deliberations in terms of seasonal climatic conditions (e.g., snow and ice) (Ripat, Brown, and Ethans 2015; Ripat and Colatruglio 2016); funding availability (e.g., one device every five years); condition trajectory (progressive vs. stable); or functional variability (e.g., diurnal variation in fatigue with MS).
Six principles were identified by WHO that need to be considered to meet the obligations for AT provision described in the Convention on the Rights of Persons with Disabilities. These are acceptability (e.g., efficiency, reliability, simplicity, safety, and aesthetics); accessibility; adaptability (i.e., can the AT be customized to meet individual needs?); affordability; availability; and quality. All six are certainly applicable to the use of exoskeletons as an AT for ambulation or manipulation. In terms of acceptability, safety is perhaps the key concern (Wolff et al. 2014). Although concerns about the potential for injury with exoskeletons frequently focus on the user, it is also important to consider the potential for injury to those around them. Reliability is also related to safety. If people begin to depend on exoskeletons as assistive devices, equipment breakdowns could become potentially dangerous (e.g., being stuck outside and unable to move in subzero temperatures). Speed is also likely factored into acceptability. Slow gait speeds might be a source of frustration for users or potentially a safety concern for activities such as street crossing. Effort is also a consideration, so that fatigue from ambulation using an exoskeleton does not limit capacity to perform other activities. Training requirements are critical; if the use of an AT is not second nature, it cannot be fully integrated into a person’s daily routines (i.e., habituation) (Mortenson et al. 2012).
Some people may only be willing to use an AT if they can do so independently. Ease of use is an important aspect for users, as they would prefer devices that are easy to doff and don and do not interfere with other activities, such as transfers. Similarly, exoskeletons should be easy to control (i.e., not difficult to program) and not be cognitively taxing.
Although exoskeletons might be considered a means of improving accessibility, there may be limitations in terms of the kinds of surfaces they can operate on, inclines they can go up and down, height of curbs and stairs they can negotiate, and transportation requirements (e.g., Could the user get in and out of a standard passenger vehicle? Would the user need to sit down when using public transportation?).
Any technology device will be poorly regarded if it can only be used during specific “windows of control” when the system is capable of being controlled. This is called “synchronous control” and plagues other advanced experimental technology, such as brain-computer interfaces (BCIs; Borisoff, Mason, and Birch 2006). Thus, the ability for exoskeletons to be operated under asynchronous control, a more natural and preferred control scheme, will be important. This would allow users to be able to operate an exoskeleton whenever they would like. This also likely affects a rarely considered characteristic of AT: spontaneity. Spontaneity represents the integration of many of the elements listed previously (e.g., ease of use, donning and doffing, and asynchronous control) and gets at the concept of transformative technology, which is at the heart of future innovations.
When considering these characteristics of effective AT devices, it quickly becomes apparent that exoskeletons are lacking at this time. A more instructive approach may be to question the short-term feasibility of exoskeletons as mobility or manipulation devices and provide some guidance about the necessary features needed to be considered for a successful design to be realized.