External Devices in Rehabilitation Context: Comparison
View Latest Version
Please note this is a comparison between Version 2 by Jason Zhu and Version 1 by António Diogo André.

Human societies have been trying to mitigate the suffering of individuals with physical impairments, with a special effort in the last century. In the 1950s, a new concept arose, finding similarities between animal exoskeletons, and with the goal of medically aiding human movement (for rehabilitation applications). There have been several studies on using exosuits with this purpose in mind. 

  • external device
  • biomechanical design
  • structural materials
  • actuation
  • energy sources
  • control system

1. Introduction

After a trauma or surgical procedure, continuous passive motion devices are typically used in rehabilitation to reduce edema, bleeding, pain, and inflammation. These devices are the first step in the rehabilitation process. Active assistive movement is also used, which helps the patient perform desired movements with the help of a suit that assists in completing the movement. In cases of neurological rehabilitation, this method is the first choice to stimulate neuroplasticity and reduce common side effects such as muscle weakness. Active resistive motion involves applying an external resistive force against a dynamic or static muscle contraction and is an effective way to increase bone and muscle mass, making it essential for musculoskeletal rehabilitation. Exoskeleton usage can enhance the results of different physiotherapeutic approaches. Still, the final outcome depends on a range of rehabilitation factors, including timing, intensity, repetition, frequency, and task-specific training protocols [33][1].
Wearing an external device, such as an exoskeleton, can provide numerous advantages in a medical rehabilitation environment [34][2], not only for the patients but also for clinical centers. These devices can enable patients to perform intensive and repetitive movements with precision, minimizing the physiotherapist’s intervention [35][3]. This can relieve therapists from fatigue and constant attention requirements. Additionally, this kind of technology can enable the rehabilitation of patients at their homes via video conference. Exoskeletons can also be used to evaluate recovery levels by measuring force levels and movement patterns [8][4]. This data can be collected from sensors [36][5] in the device itself and/or from motion capture devices that track motion patterns. This training can help people relearn lost motor functions and perform daily tasks.
The drawbacks of existing solutions should be the object of careful consideration, taking into account the person and their particular circumstances. For example, some solutions may not be energy efficient, leading to high energy consumption [26][6], while others may make it difficult for the user to interact effectively with their surroundings [37][7].
When rehabilitating a patient using an exoskeleton, the need for a large empty room must be taken into account. Moreover, since a one-size exoskeleton cannot accommodate all users due to differences in body proportions, the creation of an adjustable device that can fit all sizes poses a great challenge due to its complexity. Thus, a disproportional device regarding the body may have a negative psychological impact on the user, leading to some reluctance to make use of it [38][8].
Despite all the challenges, researchers have already developed reliable solutions to rehabilitate or enhance various parts of the body, such as ankles [39,40[9][10][11],41], hands [33[1][12],42], shoulders [43][13], lower limbs [44][14], upper limbs [45][15], arms [46][16], and back [47,48][17][18].

2. Mechanical Design

Design is an imperative aspect to consider in exoskeleton development, as every detail affects the user experience, with the final appearance being the first overall impression. During conception, several design considerations come into play during all stages of the project, from selecting structural materials to selecting control systems, with particular attention paid to key components such as batteries. An intelligent arrangement of actuators and energy sources (e.g., batteries) brings benefits beyond just aesthetics; it can improve weight distribution [27,49][19][20] and in some cases even reduce power consumption [50][21], which is directly linked to the choice of power source. Most importantly, a good design can make a positive first impression on the users, providing them with a sense of comfort, ergonomics, confidence, and convenience.
In addition to visual appearance, it is crucial to consider the technical aspects when designing the final solution. The movement’s kinematic and dynamic degrees of freedom (DOF) found in the human body [51][22] based on anthropometry should be present, as a concept, throughout the projects. The range of motion, joint torque requirements, joint rotational velocity, and joint angular bandwidth [52][23] must also be factored in. The developed device aims to aid and follow human movement without constraint or interference with the natural freedom of movement [53][24].
Creating and implementing practical solutions can be a challenging task due to the inherent complexity of the principles involved and their combination. A specific example of this complexity can be observed in the development of complete limb external skeletons. These particular devices are capable of an infinite number of combined movements, as they rely on seven distinct DOF [50][21] positioned along the limbs. These DOF are vital for daily activities [55][25], with lower limbs having three DOF at the hip, one at the knee, and three at the ankle, while upper limbs possess three at the shoulder (abduction-adduction, flexion-extension, and internal-external rotation), one at the elbow, one at the forearm, and two at the wrist.
Providing the necessary DOF for full-body applications becomes challenging with traditional exoskeletons, which often consist of rigid materials assembled in a series of fixed links. Their non-flexible characteristics can lead to problems of hyperstaticity [56][26] and can result in increased device complexity, which further complicates the design process. As an alternative, soft structures composed of mechanisms without rigid components, featuring elastic or elastomeric materials with softer (more flexible) mechanical properties, have emerged. As demonstrated by successful lightweight and flexible designs [57][27] and greater adaptability to both movement and the human body [58][28], they offer a promising alternative to their traditional rigid counterparts.
The primary function of the wearable device is not just to track human movement, but also to provide assistance by generating the necessary force or moment to hold the joints (e.g., elbow) in certain positions during daily activities or rehabilitation. Moreover, as per [59][29] guidelines, the device should be capable of generating the appropriate amount of auxiliary force or momentum to perform those daily tasks. However, it may be impossible to devise a solution that combines all the necessary DOF with adequate motion generation, as illustrated by the challenge of creating a wearable finger device. This limb is essential for performing basic daily tasks, such as typing or writing, and any solution must be practical and effective in addressing these needs. For example, opening a jar with a finger wearable device would require up to 120 N of force and 3 Nm of torque on the metacarpophalangeal (MCP) joint [60][30], without neglecting other considerations such as overall aesthetic and having four DOF [61,62][31][32].
The project must prioritize security measures, as an equal fundamental design factor. Following these safety standards avoid accidents and ensure the user’s protection in unforeseen events, like power loss or current leaks. The probability of accidents using exoskeletons is real and remains a significant concern, since estimations suggest around 4 out of every 100 users may encounter issues [63][33]. The overall solution’s appearance, functionality, safety, and ease of use are determined by the final design concept, which cannot disconnect design choices from fundamental design variables and options. Consequently, the final solution must represent a balance between design options, structural materials, actuators, energy sources, and control systems to achieve the best overall solution.

3. Structural Materials

While the terms exoskeleton and exosuit have often been used interchangeably, some argue that exosuit is a more accurate description for these devices even though the general public is more acquainted with the term exoskeleton. Despite their similarities, those terms are not synonyms when the context involves structural materials. In reality, these words represent two distinct approaches to solving the same problem. Exoskeletons are typically constructed of rigid and metallic components [64[34][35],65], while exosuits are designed using soft and flexible materials [66,67][36][37]. Although they are classified differently, both solutions should be investigated together as they offer complementary features [68][38]. Regardless of the type of material used, it is essential that any solution designed for use in rehabilitation settings meets certain critical requirements that ensure safety, as mentioned in a study by Xiloyannis [49][20]. In particular, mechanical properties assume great importance since patients undergoing rehabilitation are often susceptible to minor accidents, such as small falls, and the material should be able to withstand and resist fatigue, as pointed out by Bogue [69][39]. These characteristics are vital for ensuring that the device has a long lifespan, even when deployed on a higher number of patients during rehabilitation. Additionally, the material should offer a warm and comfortable sensation to the wearer.

Rigid vs. Soft Materials

When it comes to a rigid approach, materials like stainless steel [69][39], aluminum [70][40], and titanium [65][35] are widely used. The final solution can involve one or multiple materials for example, with frames made from aluminum and joints made from stainless steel or titanium. This multi-material approach can offer several benefits, such as reducing weight and increasing mechanical strength at critical joints. In fact, using multiple materials is becoming popular in engineering because it provides a better balance between performance, cost, and durability.
Compared to exosuits, more rigid solutions offer some advantages but bring some disadvantages. Exoskeletons offer increased mechanical strength, making them an ideal solution when high levels of torque and strength are required. In fact, these devices can withstand up to 1 GPa of tension before experiencing plastic deformation and can endure up to 50% of strain before reaching a breakdown point [71][41]. Such impressive performance metrics highlight the potential benefits of using exoskeletons in various settings.
However, the materials used are typically heavier, which can limit their portability and cause discomfort for the user [72][42]. Additionally, achieving perfect alignment between the device and the user’s joints can be a challenge, resulting in larger inertial loads that can lead to abnormal motion patterns [73][43]. Other common problems associated with rigid solutions include reduced usability and poor aesthetics, as noted by several authors [74,75][44][45]. Despite these drawbacks, rigid solutions remain popular in many applications due to their mechanical reliability and stability.
While exosuits and exoskeletons share some common characteristics, such as safety features (when applied to these devices) and price range, there are clear differences in their design and construction. Exosuits typically have symmetric properties not being susceptible to misalignment, largely due to the materials used in their production. These materials primarily consist of polymeric or composite materials, including elastomers such as liquid crystal, dielectric, and acrylic elastomers [66[36][38],68], shape memory polymers (SMPs) such as those based on epoxy and polycaprolactone materials [66][36], electroactive polymers (EAPs) like polyvinylidene difluoride (PVDF) [76[46][47],77], and conducting polymers such as polypyrrole [78,79][48][49]. Their use in exosuits allows for greater flexibility and symmetry compared to their rigid exoskeleton counterparts.
Composite materials can be comprised of metallic and polymeric substances combined with carbon fibers [57,69][27][39]. In some other cases, a solution made from chloroprene and polyurethane (PU) may also be used [57][27]. Additionally, textiles may also be utilized for certain applications [73,79][43][49]. By combining these materials, it becomes possible to create lightweight and durable devices that can provide users with a wide range of benefits.
In general, these materials enable movement smoothness [80][50], comfort, portability, flexibility, lightweight (low density) [68[38][43],73], adaptation to bioorganisms [66][36] and even the ability to emulate biological muscles [77][47]. Some of these materials can exceed their structural role and be used as actuators [77][47] since they are prone to deformations with associated large volume changes in response to external stimuli [66][36].

4. Actuators and Energy Sources

Actuators play a critical role in wearable external devices, facilitating human movement by powering them, and in this way, enabling a better interaction with the surrounding environment. In a medical context, they can be particularly valuable for helping patients undergoing rehabilitation by providing controlled motion patterns. As such, actuators are an indispensable component of many modern wearable devices, and their effectiveness can have a significant impact on user outcomes.
Actuators can be classified as either powered or unpowered, resulting in the creation of either active or passive external devices [81][51], respectively. Powered alternatives may be noisier and are generally costlier due to the need for additional components, as well as requiring users to carry bulky energy-supply systems [82][52]. On the other hand, passive devices do not require power units, making them lighter and weighing up to a fourth of their powered counterparts. A good example of this is the ankle exoskeleton developed by Mooney et al. [83][53] and Collins et al. [84][54], which aims to reduce the metabolic rate during walking [85][55].

4.1. Traditional Actuators

Traditional actuators typically are based on rigid systems, allowing them to generate higher forces [49][20], greater movement precision, and improved dynamic performance [68][38], and as a result, making them ideal for more complex tasks such as severe mobility disorders. However, it brings some disadvantages, such as leading to higher power consumption [68][38]. When a power supply is required to input the actuator, the user’s freedom of movement can be limited. Additionally, elderly users may feel uncomfortable with the robotic aspect of the actuators, which can convey a detached and cold sensation and lead to their refusal to use the device.
Purely mechanical actuators such as springs [86][56], are commonly used in unpowered devices (which do not require any external source of energy) and convert the tension force from the actuators into torques at the joints [44][14]. This mechanical solution can help to reduce the metabolic consumption of energy [84][54] during walking or running activities [87][57]. However, the usability of such actuators has limited usefulness in rehabilitation cases, as they only provide passive assistance. For example, during walking, the user must first tense the actuator during flexion movement in order to receive assistance in the extension movement.
Mechanical servomotor-based actuators [88][58] are a simple and direct approach for achieving actuation through electrical stimulation. They provide motion and assistance when connected to the structural material (soft or rigid). However, due to the nature of the input type, they always require an external source of electrical energy, such as (portable) batteries. Plus, they are also rigid and bulky, which can limit the flexibility of the entire system [89][59].
Pneumatic-based actuators are a highly efficient and safe solution in terms of linear and rotational movement control since the actuator’s motion is converted from pressurized air energy [89][59]. Also, they are particularly suitable for applications that demand repetitive opening and closing tasks, as well as in environments of extreme temperatures or even in industrial applications where other types of actuators are not viable alternatives. As air-compressed-based actuators, this type of solution can convert up to 6 bar of pressure into movement, if necessary. However, to perform all of this and enable movement, connectivity to a rigid control and power system, such as a compressor, is a mandatory aspect requirement [89][59], which can occasionally lead to pressure drops and noise. Moreover, pneumatic actuators could be produced either considering rigid [90][60] or soft materials, such as latex or rubber tubes [91][61], which make them a feasible solution for exoskeletons [90][60] and exosuits [91][61].
Hydraulic actuators [92,93][62][63] share similar advantages and disadvantages when compared to pneumatic actuators. Similarly, they require a hydraulic fluid to output linear, rotary, or even oscillatory movements by the actuator, but as liquids are nearly incompressible, the force produced is considerably higher. The exoskeleton/exosuit movement is thus achieved by converting hydraulic into mechanical energy.

4.2. Soft Actuators

Actuation solutions based on soft actuators can be a comfortable alternative when used during the rehabilitation process [119][64] and unlike the traditional methods, they can be stimulated externally by different inputs. The direct incidence of light, heat, electric or magnetic fields results in mechanical movement performed by the actuators [77,89][47][59]. They can be thus defined as mechanical and electrical elements whose output/operation varies under different physical, chemical, and/or biological stimuli. Typically, these soft actuators can be built using a different range of materials, from particles to polymers, such as EAPs [77][47] or SMPs [120][65], papers [89][59], fluids, shape memory alloys (SMAs) [89][59], hydrogels, liquid materials [66][36], 2D materials, carbon-based materials [66][36] or combinations thereof [89][59]. Despite the numerous alternatives, not all of these soft actuators are viable for rehabilitation cases. The pertinence of their applicability is based on performance parameters such as stress, strain, Young’s modulus, power, energy, and force density [89][59].
Electrically responsive soft materials are flexible and stretchable materials able to convert external electric inputs into mechanical response outputs. Depending on the type of material, they are classified as dielectric elastomer actuators (DEAs), piezoelectric-based actuators, and electrically conducting polymers (ECPs). DEAs have their input-output conversion based on Coulombic attraction. Two flexible electrodes with a potential difference located on separate ends of a compressible membrane are used to obtain the mechanical response from DEAs [89][59]. They are highly flexible materials with high energy density, strains, and the ability to emulate the behavior of biological muscles [94][66]. The performance of these materials depends on their stability, breakdown voltage, and dielectric constant of them [89][59]. However, they generally require high voltages, usually in the kV range, to perform and leakage currents are often observed when high electric fields are applied, especially when the actuator ages [89][59]. Adding liquid elastomers to DEAs has been proposed as a solution for these limitations [89][59]. Examples of dielectric materials found in the literature include acrylic elastomers [95][67], which are highly deformable and possess high viscoelasticity. However, the actuator’s bandwidth could be limited due to these mechanical properties [89][59]. Other examples include silicone-based materials [96][68] and PU-based elastomers [97][69]. The PU-based elastomers have faster reactions and can be cast into various shapes, but they perform significantly lower strains than the dielectric materials [89][59]. DEA solutions are constantly being researched and developed to enhance their properties in the actuation field [98][70].
Piezoelectric-based actuators are capable of producing voltage or electric charge in the presence of mechanical or vibrational forces (direct effect) or deformation when electrically-stimulated (indirect effect) [77,99][47][71]. These actuators can operate in room conditions for long periods and have a quick response time, typically in the milliseconds range. Also, they can hold strain under activation, inducing relatively large actuation forces [77][47]. However, their usability in real-world scenarios can be limited by the large AC voltages required [89][59]. Common piezoelectric materials for actuation and sensors include PVDF and its copolymers [100[72][73],101], graphene [102][74], and zirconate titanate [103][75], among many others [104][76]. ECPs [105][77] are organic polymeric materials obtained by reduction or oxidation reactions [106][78]. They can conduct electricity with conductivities up to 105 S/cm, achieved through traditional sources, such as batteries or chemical reactions. Moreover, these electrically responsive material types have been powered using biofuels, such as glucose, which shows their potential as an environmentally friendly source of energy [107][79]. Polypyrrole, a type of ECP obtained by the oxidative polymerization of pyrrole, is characterized by high mechanical properties and chemical stability [106][78] and has been shown to emulate human biological muscles due to its similar behavior and low voltage operability [107,108][79][80]. These characteristics make ECPs an interesting choice due to their biomimetic and biocompatible nature.
Magnetic responsive materials have potential applications as actuators since they are easily controllable through magnetic field direction and magnitude, which can penetrate most materials [89][59]. This feature makes them a promising solution for use in restricted or enclosed areas [109][81]. This actuation method is based on incorporating magnetic particles and fillers into different soft compounds such as polymers, gels, papers, or fluids [109][81]. This results in a magnetization profile with variable magnitude and direction [110][82]. In the presence of a magnetic field, the particles or fillers align to create deformation, bending, elongation, or contraction [89][59]. These magnetic-based actuators have a fast response time, with literature reporting speeds of up to 100 Hz [111][83]. However, there are some disadvantages associated with the magnetic coils used to generate magnetic fields. Their large size, high energy consumption and limited control areas where the magnetic field may not be strong enough are some handicaps to consider [89][59].
Thermally responsive materials, including silicone-based elastomer materials [116][84], liquid crystal elastomers, and synthetic hydrogels [89][59], can be activated by a thermal source, such as infrared (IR) radiation, thermal radiation, or Joule heating [113][85]. For instance, shape-memory materials (SMM) [115][86] can be deformed by external forces and return to their original “memorized” shape under loading or thermal cycles [89][59]. These materials include SMAs (typically iron-based or copper-based) [114][87], which return to their original shape when the temperature exceeds a certain threshold after deformation, and SMP materials (PU and thermoplastic PU) [112][88]. SMPs are cost-effective, have high elastic deformation, and are easy to manufacture [89][59]. Furthermore, they can be activated remotely, for instance, through laser incidence, and are often safer than electrical fields for biomedical applications [89][59]. Some of these light actuators are capable of lifting objects that are up to 200 times heavier than their own weight, to up to 5 mm height [113][85]. However, such thermally responsive materials tend to have slower response times and are less efficient compared to other types of stimuli-based actuators [89][59].
Photo-responsive materials employ photochromic molecules to capture optical signals and convert them into property modifications [117,118][89][90]. They represent an attractive wireless alternative, as they can be controlled in small sizes and consume low energy [89][59]. However, slow actuation speed and mechanical property degradation remain major limitations [89][59]. Photochromic molecules, such as spiropyran [117][89], may be added to various materials, such as gels, polymers, and fluids, to render them photoresponsive [89][59]. They respond to the light spectrum, visible or near-IR) [89][59].

5. Control

5.1. Control System Architectures

The control system of external skeleton or suit devices can be categorized into four main architectures—model-based, hierarchy-based, physical parameters-based, and usage-based [121][91]. While none of these architectures have been used individually due to their complexity or effectiveness, they are frequently combined to achieve the desired control of a specific device [121][91].
In general, model-based control systems can be further classified into two types—dynamic and muscular models [124][92]. The dynamic model reflects the human motion intent by combining inertial, gravitational, Coriolis, and centrifugal effects to model the human body as a series of rigid links connected by joints (bones) [125,126][93][94]. The control system of BLEEX is just an example of a dynamic model-based system [50][21]. This based-type architecture is even developed through different approaches: mathematical, system identification, and artificial intelligence models. To obtain a mathematical architecture for the external device based on the physical characteristics of the system, the system requires a precise dynamic model [121][91]. For instances in which a dynamic model cannot be adequately developed through theoretical mathematical models, the system identification model is often utilized [127][95]. The artificial intelligence method is the most popular approach to identifying the dynamic model due to its efficiency [127][95].
Muscle-based models have also been utilized in exoskeleton control systems. Unlike dynamic models, these models predict the muscle forces generated by human joints as a function of muscle neural activities and joint kinematics [126,128][94][96]. This approach, which can be obtained by using parametric or non-parametric models, takes the electrical signal produced by muscles as input and sends force estimation as output to actuators [121][91]. The parametric muscle model is commonly implemented using the Hill-based model, which refers to muscle contraction and uses the estimated muscle activation level [129,130,131][97][98][99]. It is comprised of three elements: a contractile element, representing force generated by active muscle fibers, a series element, which models the mechanical response of the muscle, and a parallel element which simulates the passive resistance of muscles to stretch [131][99]. In addition, the output sent from this type of control model is a function of electromyographic (EMG) neural activity and muscle length [121][91]. In contrast, non-parametric muscle models do not require knowledge about muscle and joint dynamics but they can be the source of control inefficiencies [132][100] (ex. finite impulse response model).
Shafer et al. [133][101] developed an ankle exoskeleton controller that uses a control system based on a neuromuscular model. They make conclusions on the effectiveness of their model in providing a wide range of assistance torque and power. Moreover, Song et al. [134][102] developed a novel model-based control to predict motion trajectories and amplify the forces produced by the user.
The hierarchy-based control system, exemplified in Huang et al. [135][103] and Dinh et al. [136][104], utilizes a hierarchical structure to manage inputs and outputs. The controllers are divided into three levels: task level, high level, and low level. The task-level controller, which is the highest level, is responsible for performing the designated tasks [121][91]. The high-level controller adjusts the force of human-external device interaction based on information received from the task-level controller [121][91]. Finally, the low-level controller is responsible for controlling the position and/or force performed by the exoskeleton joints, therefore contacting directly to the exosuit [121][91].
Copaci et al. [137][105] implemented a hierarchy-based control system in an elbow exoskeleton. Using algorithms to process EMG signals, they were capable of generating position and torque references in SMA actuators used for active rehabilitation therapies.
Control strategies such as those utilized in the ARMin [138][106], RUPERT IV [139][107], and LOPES [140][108] exoskeletons use physical parameters as a basis for their implementation. These solutions can be classified as either position, torque/force, or force interaction controllers [121][91]. The low-level controller in the position control scheme ensures that the exoskeleton joints turn to the desired angle, while the torque/force controller regulates the desired force and/or torque [141][109], and is also classified as a low-level controller [121][91].
The interaction force controller, typically functioning as a high-level controller, is responsible for providing appropriate assistance to users during a task [121][91]. This physical parameter controller takes into consideration the force interaction between the user and the exoskeleton, which is considered in an external device [121][91]. The impedance controller, which accepts position and produces force, or the admittance controller, which accepts force and yields position, can be used to control this physical parameter controller [142][110].
The impedance controller is typically more effective for lightweight, backdrivable external devices (such as cable-driven devices) compared to other controllers [124][92]. It extends the position control, enabling it to not only regulate the position and force but also the relationship and interaction between the exosuit and the human body [142,143][110][111]. This controller architecture includes an impedance module, which receives the error position of the joints and yields the force values that serve as force references for subsequent stages. The architecture also comprises a force/torque controller that attempts to ensure that the forces exerted by the exoskeleton actuators are approximately equal to force references [121][91].
The admittance controller is employed to regulate the force generated by the external skeleton during interaction with the user [144][112]. It features an admittance model, which receives forces and outputs the position, as well as a position controller that controls the joint angle based on position references from the admittance model output [121][91].
Wu et al. [145][113] implemented a physical parameter-based control system in an exoskeleton for upper-limb rehabilitation of disabled patients. They used a modified sliding mode control strategy incorporating a proportional integral derivative (PID) sliding surface and a fuzzy hitting control law to ensure a robust and optimal position control performance. Their approach led to the best control performances in terms of tracking accuracy, response speed, and robustness against external disturbances.
The usage-based control systems, such as those implemented in MGA [146][114] and L-Exos [147][115], can be categorized into three types: virtual reality (VR) controller, teleoperation controller [121][91], and gait controller, which is commonly used in lower limb solutions [140][108]. VR controllers are commonly employed in rehabilitation exercises for upper-limb exoskeletons [148][116]. They allow for the guidance and assistance of patients during tasks such as moving a virtual object with their hands [139][107], virtually painting a wall [146][114], or carrying out constrained motion tasks [147][115]. In these applications, the exoskeleton/exosuit can be regarded as a haptic device [121][91].
The teleoperation controller is a form of master-slave controller, where the exoskeleton worn by the user is commonly used as the master type and a mirror robot serves as the slave [121][91]. In this configuration, interaction control occurs between the slave robot and the environment, as opposed to the typical interaction between the user and the exoskeleton [121][91]. Rahman et al. [149][117] implemented a teleoperation controller in an exoskeleton for rehabilitation and passive arm movement assistance (MARSE-4), constituted by an upper-limb prototype and a master exoskeleton arm (mExoArm). While mExoArm is operated by the patient, the upper-limb prototype mirrors the movement.

5.2. Sensors

Capturing human motion intents for external device control is a major challenge, which can be addressed through the use of sensors associated with both the control system and the device [151][118]. These sensors capture the user’s movement intention as an input signal to the control system, which then provides output to the exosuit to perform the intended move. To ensure success, this input signal must be precise and accurate. In addition to the intention-prediction instrumentation, other sensors such as inertial measurement units [152][119] (e.g., gyroscopes [153][120] and accelerometers [154][121]) or mechanical sensors [155][122] can be employed to measure or evaluate the output movement. However, it should be noted that these sensors are unable to predict movement beforehand [156][123].
Several control methods have been proposed to detect human intention through human–robot interaction dynamics, which could effectively assist able-bodied human subjects [157,158][124][125]. While control methods using human–robot interaction dynamics are effective in assisting able-bodied humans, they may not always be suitable as the user needs to produce sufficient torque at joints to initiate movement. If this amount of torque is not generated, the device may not be effectively controlled, resulting in a problematic aspect for elderly or severely disabled individuals [159][126]. The ideal solution for human–robot interaction entails the prediction of movement intention, instead of a reaction to a precursor movement. This approach can improve performance in scenarios where generating sufficient torque is not possible [122][127].
To predict human movement, electrophysiological signals from proteins, organs, or muscles can be captured through sensors measuring voltage changes or electric current [160,161][128][129]. EMG sensors (intramuscular [162][130], surface [163][131]) can measure small electrical signals [164][132] produced by muscle contraction and have been successfully used in exoskeleton control [165,166][133][134]. EMG-based methods can capture the user’s intention to control the device, even if the person cannot produce sufficient joint torques or execute a particular movement [122][127]. However, the signal measured by EMG sensors might be biased by various factors, such as muscle crosstalk susceptibility [151][118], skin condition (surface sensors), muscle fatigue [156][123], or the inaccessibility of deep muscle fibers [167][135].
In addition to the use of EMG sensors, there are other sensors that can be considered to be alternatives for measuring muscle electrical activity. One such alternative is mechanomyography (MMG) sensors which are less sensitive to skin conditions compared to EMG sensors [156,168][123][136]. These sensors measure the signal produced by muscles with respect to the gross lateral muscle movements which causes low-frequency vibration during contraction, lateral vibrations at the muscle’s resonant frequency, and volumes introduced by the changes in the muscles [169][137]. Despite the advantages, MMG sensors have some disadvantages, such as being affected by muscle fatigue as well [156][123]. Sonomyography (SMG) sets up another possibility to predict the user’s movement intention by measuring muscle thickness and tracking skeletal muscle deformation from superficial to deep tissue [170,171][138][139]. SMG sensors are also capable of classifying several motions and predicting joint kinetics during dynamic activities, such as those in the wrist [171,172][139][140]. However, muscle fatigue is still a common issue with SMG sensors [156][123].
Finally, Electroencephalogram (EEG) sensors can capture the user’s intention without using sensors that measure the signal produced directly in muscles [156,174][123][141]. Instead, they measure the electrical activity in the brain. However, the signal captured by EEG sensors is not accurate enough and can only be used for classifying movements [156][123].

References

  1. Haghshenas-Jaryani, M.; Patterson, R.M.; Bugnariu, N.; Wijesundara, M.B. A pilot study on the design and validation of a hybrid exoskeleton robotic device for hand rehabilitation. J. Hand Ther. 2020, 33, 198–208.
  2. Iqbal, J.; Tsagarakis, N.G.; Caldwell, D.G. A human hand compatible optimised exoskeleton system. In Proceedings of the 2010 IEEE International Conference on Robotics and Biomimetics, Tianjin, China, 14–18 December 2010.
  3. Iqbal, J.; Baizid, K. Stroke rehabilitation using exoskeleton-based robotic exercisers: Mini Review. Biomed. Res. 2015, 26, 197–201.
  4. Chen, B.; Ma, H.; Qin, L.Y.; Gao, F.; Chan, K.M.; Law, S.W.; Qin, L.; Liao, W.H. Recent developments and challenges of lower extremity exoskeletons. J. Orthop. Transl. 2016, 5, 26–37.
  5. Porciuncula, F.; Roto, A.V.; Kumar, D.; Davis, I.; Roy, S.; Walsh, C.J.; Awad, L.N. Wearable Movement Sensors for Rehabilitation: A Focused Review of Technological and Clinical Advances. PM&R 2018, 10, S220–S232.
  6. van den Bogert, A.J. Exotendons for assistance of human locomotion. BioMed. Eng. OnLine 2003, 2, 17.
  7. Baser, O.; Kizilhan, H.; Kilic, E. Biomimetic compliant lower limb exoskeleton (BioComEx) and its experimental evaluation. J. Braz. Soc. Mech. Sci. Eng. 2019, 41, 226.
  8. Mihaleva, G.; Kow, F.C.S.F. Development of an Aesthetic for a Stroke Rehabilitation System. In Proceedings of the 2021 International Symposium on Wearable Computers, Virtual Event, 21–26 September 2021.
  9. Bae, J.; Siviy, C.; Rouleau, M.; Menard, N.; Odonnell, K.; Geliana, I.; Athanassiu, M.; Ryan, D.; Bibeau, C.; Sloot, L.; et al. A Lightweight and Efficient Portable Soft Exosuit for Paretic Ankle Assistance in Walking After Stroke. In Proceedings of the 2018 IEEE International Conference on Robotics and Automation (ICRA), Brisbane, Australia, 21–26 May 2018.
  10. Asbeck, A.T.; Dyer, R.J.; Larusson, A.F.; Walsh, C.J. Biologically-inspired soft exosuit. In Proceedings of the 2013 IEEE 13th International Conference on Rehabilitation Robotics (ICORR), Seattle, DC, USA, 24–26 June 2013.
  11. Grimmer, M.; Quinlivan, B.T.; Lee, S.; Malcolm, P.; Rossi, D.M.; Siviy, C.; Walsh, C.J. Comparison of the human-exosuit interaction using ankle moment and ankle positive power inspired walking assistance. J. Biomech. 2019, 83, 76–84.
  12. Aggogeri, F.; Mikolajczyk, T.; O’Kane, J. Robotics for rehabilitation of hand movement in stroke survivors. Adv. Mech. Eng. 2019, 11.
  13. O’Neill, C.T.; Phipps, N.S.; Cappello, L.; Paganoni, S.; Walsh, C.J. A soft wearable robot for the shoulder: Design, characterization, and preliminary testing. In Proceedings of the 2017 International Conference on Rehabilitation Robotics (ICORR), London, UK, 17–20 July 2017.
  14. Zhou, L.; Chen, W.; Chen, W.; Bai, S.; Zhang, J.; Wang, J. Design of a passive lower limb exoskeleton for walking assistance with gravity compensation. Mech. Mach. Theory 2020, 150, 103840.
  15. Gopura, R.A.R.C.; Kiguchi, K.; Li, Y. SUEFUL-7: A 7DOF upper-limb exoskeleton robot with muscle-model-oriented EMG-based control. In Proceedings of the 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, St. Louis, MO, USA, 10–15 October 2009.
  16. Garcia, D.; Soto, V.; Lurbe, Y.; Morris, M.; Tosunoglu, S. Development of a rehabilitative exoskeletal arm. ASME Early Career Tech. J. 2012, 11, 11–17.
  17. Hensel, R.; Keil, M. Subjective Evaluation of a Passive Industrial Exoskeleton for Lower-back Support: A Field Study in the Automotive Sector. IISE Trans. Occup. Ergon. Hum. Factors 2019, 7, 213–221.
  18. Koopman, A.S.; Kingma, I.; de Looze, M.P.; van Dieën, J.H. Effects of a passive back exoskeleton on the mechanical loading of the low-back during symmetric lifting. J. Biomech. 2020, 102, 109486.
  19. Sado, F.; Yap, H.J.; Ghazilla, R.A.R.; Ahmad, N. Design and control of a wearable lower-body exoskeleton for squatting and walking assistance in manual handling works. Mechatronics 2019, 63, 102272.
  20. Xiloyannis, M.; Cappello, L.; Binh, K.D.; Antuvan, C.W.; Masia, L. Preliminary design and control of a soft exosuit for assisting elbow movements and hand grasping in activities of daily living. J. Rehabil. Assist. Technol. Eng. 2017.
  21. Zoss, A.; Kazerooni, H.; Chu, A. Biomechanical design of the Berkeley lower extremity exoskeleton (BLEEX). IEEE/ASME Trans. Mechatron. 2006, 11, 128–138.
  22. Hansen, C.; Gosselin, F.; Mansour, K.B.; Devos, P.; Marin, F. Design-validation of a hand exoskeleton using musculoskeletal modeling. Appl. Ergon. 2018, 68, 283–288.
  23. Cenciarini, M.; Dollar, A.M. Biomechanical considerations in the design of lower limb exoskeletons. In Proceedings of the 2011 IEEE International Conference on Rehabilitation Robotics, Zurich, Switzerland, 29 June–1 July 2011.
  24. Privitera, M.B.; Evans, M.; Southee, D. Human factors in the design of medical devices—Approaches to meeting international standards in the European Union and USA. Appl. Ergon. 2017, 59, 251–263.
  25. Rosen, J.; Perry, J.; Manning, N.; Burns, S.; Hannaford, B. The human arm kinematics and dynamics during daily activities—Toward a 7 DOF upper limb powered exoskeleton. In Proceedings of the ICAR 05, 12th International Conference on Advanced Robotics, Seattle, WA, USA, 18–20 July 2005.
  26. Jarrasse, N.; Morel, G. A formal method for avoiding hyperstaticity when connecting an exoskeleton to a human member. In Proceedings of the 2010 IEEE International Conference on Robotics and Automation, Anchorage, AK, USA, 3–7 May 2010.
  27. Thalman, C.M.; Lee, H. Design and Validation of a Soft Robotic Ankle-Foot Orthosis (SR-AFO) Exosuit for Inversion and Eversion Ankle Support. In Proceedings of the 2020 IEEE International Conference on Robotics and Automation (ICRA), Paris, France, 31 May–4 June 2020.
  28. Rus, D.; Tolley, M.T. Design, fabrication and control of soft robots. Nature 2015, 521, 467–475.
  29. Zatsiorsky, V.M.; Gregory, R.W.; Latash, M.L. Force and torque production in static multifinger prehension: Biomechanics and control. I. Biomechanics. Biol. Cybern. 2002, 87, 50–57.
  30. Butz, K.D.; Merrell, G.; Nauman, E.A. A biomechanical analysis of finger joint forces and stresses developed during common daily activities. Comput. Methods Biomech. Biomed. Eng. 2012, 15, 131–140.
  31. Soechting, J.F.; Flanders, M. Flexibility and Repeatability of Finger Movements during Typing: Analysis of Multiple Degrees of Freedom. J. Comput. Neurosci. 1997, 4, 29–46.
  32. Serbest, K.; Cilli, M.; Yildiz, M.Z.; Eldogan, O. Development of a human hand model for estimating joint torque using MATLAB tools. In Proceedings of the 2016 6th IEEE International Conference on Biomedical Robotics and Biomechatronics (BioRob), Singapore, 26–29 June 2016.
  33. Miller, L.; Zimmermann, A.; Herbert, W. Clinical effectiveness and safety of powered exoskeleton-assisted walking in patients with spinal cord injury: Systematic review with meta-analysis. Med. Devices Evid. Res. 2016, 2016, 455–466.
  34. Granados, D.F.P.; Kadone, H.; Suzuki, K. Unpowered Lower-Body Exoskeleton with Torso Lifting Mechanism for Supporting Sit-to-Stand Transitions. In Proceedings of the 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Madrid, Spain, 1–5 October 2018.
  35. Kagirov, I.; Kapustin, A.; Kipyatkova, I.; Klyuzhev, K.; Kudryavcev, A.; Kudryavcev, I.; Loskutov, Y.; Ryumin, D.; Karpov, A. Medical exoskeleton “Remotion” with an intelligent control system: Modeling, implementation, and testing. Simul. Model. Pract. Theory 2021, 107, 102200.
  36. Liu, J.; Gao, Y.; Lee, Y.J.; Yang, S. Responsive and Foldable Soft Materials. Trends Chem. 2020, 2, 107–122.
  37. Kushner, A.M.; Guan, Z. Modular Design in Natural and Biomimetic Soft Materials. Angew. Chem. Int. Ed. 2011, 50, 9026–9057.
  38. Chiaradia, D.; Xiloyannis, M.; Solazzi, M.; Masia, L.; Frisoli, A. Comparison of a Soft Exosuit and a Rigid Exoskeleton in an Assistive Task. In Biosystems and Biorobotics; Springer International Publishing: Cham, Switzerland, 2018; pp. 415–419.
  39. Bogue, R. Robotic exoskeletons: A review of recent progress. Ind. Robot. Int. J. 2015, 42, 5–10.
  40. Vertechy, R.; Frisoli, A.; Dettori, A.; Solazzi, M.; Bergamasco, M. Development of a new exoskeleton for upper limb rehabilitation. In Proceedings of the 2009 IEEE International Conference on Rehabilitation Robotics, Kyoto, Japan, 23–26 June 2009.
  41. Ansys. Granta Edupack Software. Available online: https://www.ansys.com/products/materials/granta-edupack/ (accessed on 1 October 2023).
  42. González-Vargas, J.; Ibáñez, J.; Contreras-Vidal, J.L.; Van der Kooij, H.; Pons, J.L. Wearable Robotics: Challenges and Trends, Proceedings of the 2nd International Symposium on Wearable Robotics, WeRob2016, Segovia, Spain, 18–21 October 2016; Springer: Cham, Switzerland, 2016; p. 449.
  43. Asbeck, A.T.; Rossi, S.M.D.; Holt, K.G.; Walsh, C.J. A biologically inspired soft exosuit for walking assistance. Int. J. Robot. Res. 2015, 34, 744–762.
  44. Almenara, M.; Cempini, M.; Gómez, C.; Cortese, M.; Martín, C.; Medina, J.; Vitiello, N.; Opisso, E. Usability test of a hand exoskeleton for activities of daily living: An example of user-centered design. Disabil. Rehabil. Assist. Technol. 2015, 12, 84–96.
  45. Hoľvová, P. Analysis of Design Today Exoskeletons in the Health Field. Appl. Mech. Mater. 2014, 613, 320–324.
  46. Correia, D.M.; Barbosa, J.C.; Costa, C.M.; Reis, P.M.; Esperança, J.M.S.S.; de Zea Bermudez, V.; Lanceros-Méndez, S. Ionic Liquid Cation Size-Dependent Electromechanical Response of Ionic Liquid/Poly(vinylidene fluoride)-Based Soft Actuators. J. Phys. Chem. C 2019, 123, 12744–12752.
  47. Bar-Cohen, Y.; Anderson, I.A. Electroactive polymer (EAP) actuators—Background review. Mech. Soft Mater. 2019, 1, 5.
  48. Ansari, R. Polypyrrole Conducting Electroactive Polymers: Synthesis and Stability Studies. e-J. Chem. 2006, 3, 186–201.
  49. Jager, E.W.; Martinez, J.G.; Zhong, Y.; Persson, N.K. Soft actuator materials for textile muscles and wearable bioelectronics. In Wearable Bioelectronics; Elsevier: Amsterdam, The Netherlands, 2020; pp. 201–218.
  50. Chiaradia, D.; Xiloyannis, M.; Antuvan, C.W.; Frisoli, A.; Masia, L. Design and embedded control of a soft elbow exosuit. In Proceedings of the 2018 IEEE International Conference on Soft Robotics (RoboSoft), Livorno, Italy, 24–28 April 2018.
  51. Roveda, L.; Savani, L.; Arlati, S.; Dinon, T.; Legnani, G.; Tosatti, L.M. Design methodology of an active back-support exoskeleton with adaptable backbone-based kinematics. Int. J. Ind. Ergon. 2020, 79, 102991.
  52. Xie, L.; Huang, G.; Huang, L.; Cai, S.; Li, X. An Unpowered Flexible Lower Limb Exoskeleton: Walking Assisting and Energy Harvesting. IEEE/ASME Trans. Mechatron. 2019, 24, 2236–2247.
  53. Mooney, L.M.; Rouse, E.J.; Herr, H.M. Autonomous exoskeleton reduces metabolic cost of human walking during load carriage. J. NeuroEng. Rehabil. 2014, 11.
  54. Collins, S.H.; Wiggin, M.B.; Sawicki, G.S. Reducing the energy cost of human walking using an unpowered exoskeleton. Nature 2015, 522, 212–215.
  55. Speakman, J.R.; Selman, C. Physical activity and resting metabolic rate. Proc. Nutr. Soc. 2003, 62, 621–634.
  56. Yandell, M.B.; Tacca, J.R.; Zelik, K.E. Design of a Low Profile, Unpowered Ankle Exoskeleton That Fits under Clothes: Overcoming Practical Barriers to Widespread Societal Adoption. IEEE Trans. Neural Syst. Rehabil. Eng. 2019, 27, 712–723.
  57. Witte, K.A.; Fiers, P.; Sheets-Singer, A.L.; Collins, S.H. Improving the energy economy of human running with powered and unpowered ankle exoskeleton assistance. Sci. Robot. 2020, 5.
  58. Ramlee, M.R.H.B.; Yusoff, H.M. Design and development of gripping assistive device for post-Stroke rehabilitation. In Proceedings of the 2017 IEEE 4th International Conference on Smart Instrumentation, Measurement and Application (ICSIMA), Putrajaya, Malasia, 28–20 November 2017.
  59. El-Atab, N.; Mishra, R.B.; Al-Modaf, F.; Joharji, L.; Alsharif, A.A.; Alamoudi, H.; Diaz, M.; Qaiser, N.; Hussain, M.M. Soft Actuators for Soft Robotic Applications: A Review. Adv. Intell. Syst. 2020, 2.
  60. Dao, D.M.; Pham, P.D.; Tran, T.X.; Le, T.T.T. Study on the transient response of lower limb rehabilitation actuator using the pneumatic cylinder. J. Mechatron. Electr. Power Veh. Technol. 2018, 9, 65–72.
  61. Al-Fahaam, H.; Davis, S.; Nefti-Meziani, S. Wrist rehabilitation exoskeleton robot based on pneumatic soft actuators. In Proceedings of the 2016 International Conference for Students on Applied Engineering (ISCAE), Newcastle upon Tyne, UK, 20–21 October 2016.
  62. Stienen, A.; Hekman, E.; ter Braak, H.; Aalsma, A.; van der Helm, F.; van der Kooij, H. Design of a Rotational Hydroelastic Actuator for a Powered Exoskeleton for Upper Limb Rehabilitation. IEEE Trans. Biomed. Eng. 2010, 57, 728–735.
  63. Glowinski, S.; Krzyzynski, T.; Bryndal, A.; Maciejewski, I. A Kinematic Model of a Humanoid Lower Limb Exoskeleton with Hydraulic Actuators. Sensors 2020, 20, 6116.
  64. Kwon, J.; Park, J.H.; Ku, S.; Jeong, Y.; Paik, N.J.; Park, Y.L. A Soft Wearable Robotic Ankle-Foot-Orthosis for Post-Stroke Patients. IEEE Robot. Autom. Lett. 2019, 4, 2547–2552.
  65. Taniguchi, H. Flexible Artificial Muscle Actuator Using Coiled Shape Memory Alloy Wires. APCBEE Procedia 2013, 7, 54–59.
  66. Qiu, Y.; Zhang, E.; Plamthottam, R.; Pei, Q. Dielectric Elastomer Artificial Muscle: Materials Innovations and Device Explorations. Acc. Chem. Res. 2019, 52, 316–325.
  67. Michel, S.; Zhang, X.Q.; Wissler, M.; Löwe, C.; Kovacs, G. A comparison between silicone and acrylic elastomers as dielectric materials in electroactive polymer actuators. Polym. Int. 2009, 59, 391–399.
  68. Madsen, F.B.; Daugaard, A.E.; Hvilsted, S.; Skov, A.L. The Current State of Silicone-Based Dielectric Elastomer Transducers. Macromol. Rapid Commun. 2016, 37, 378–413.
  69. Gallone, G.; Galantini, F.; Carpi, F. Perspectives for new dielectric elastomers with improved electromechanical actuation performance: Composites vs. blends. Polym. Int. 2010, 59, 400–406.
  70. Shintake, J.; Rosset, S.; Schubert, B.; Floreano, D.; Shea, H. Versatile Soft Grippers with Intrinsic Electroadhesion Based on Multifunctional Polymer Actuators. Adv. Mater. 2015, 28, 231–238.
  71. Gao, X.; Yang, J.; Wu, J.; Xin, X.; Li, Z.; Yuan, X.; Shen, X.; Dong, S. Piezoelectric Actuators and Motors: Materials, Designs, and Applications. Adv. Mater. Technol. 2019, 5.
  72. Costa, C.M.; Cardoso, V.F.; Brito-Pereira, R.; Martins, P.; Correia, D.M.; Correia, V.; Ribeiro, C.; Martins, P.M.; Lanceros-Méndez, S. Electroactive poly(vinylidene fluoride)-based materials: Recent progress, challenges, and opportunities. In Fascinating Fluoropolymers and Their Applications; Elsevier: Amsterdam, The Netherlands, 2020; pp. 1–43.
  73. Ren, B.; Cho, H.; Lissenden, C. A Guided Wave Sensor Enabling Simultaneous Wavenumber-Frequency Analysis for Both Lamb and Shear-Horizontal Waves. Sensors 2017, 17, 488.
  74. da Cunha Rodrigues, G.; Zelenovskiy, P.; Romanyuk, K.; Luchkin, S.; Kopelevich, Y.; Kholkin, A. Strong piezoelectricity in single-layer graphene deposited on SiO2 grating substrates. Nat. Commun. 2015, 6, 7572.
  75. Someya, T. Stretchable Electronics; Wiley-VCH Verlag GmbH: Weinheim, Germany, 2012; p. 462.
  76. Chorsi, M.T.; Curry, E.J.; Chorsi, H.T.; Das, R.; Baroody, J.; Purohit, P.K.; Ilies, H.; Nguyen, T.D. Piezoelectric Biomaterials for Sensors and Actuators. Adv. Mater. 2018, 31, 1802084.
  77. Hu, F.; Xue, Y.; Xu, J.; Lu, B. PEDOT-Based Conducting Polymer Actuators. Front. Robot. AI 2019, 6, 114.
  78. Naarmann, H. Polymers, Electrically Conducting. In Ullmann’s Encyclopedia of Industrial Chemistry; Wiley: Hoboken, NJ, USA, 2000.
  79. Mazar, F.M.; Martinez, J.G.; Tyagi, M.; Alijanianzadeh, M.; Turner, A.P.F.; Jager, E.W.H. Artificial Muscles Powered by Glucose. Adv. Mater. 2019, 31, 1901677.
  80. Maziz, A.; Concas, A.; Khaldi, A.; Stålhand, J.; Persson, N.K.; Jager, E.W.H. Knitting and weaving artificial muscles. Sci. Adv. 2017, 3, e1600327.
  81. Hines, L.; Petersen, K.; Lum, G.Z.; Sitti, M. Soft Actuators for Small-Scale Robotics. Adv. Mater. 2016, 29, 1603483.
  82. Diller, E.; Zhuang, J.; Lum, G.Z.; Edwards, M.R.; Sitti, M. Continuously distributed magnetization profile for millimeter-scale elastomeric undulatory swimming. Appl. Phys. Lett. 2014, 104, 174101.
  83. Garstecki, P.; Tierno, P.; Weibel, D.B.; Sagués, F.; Whitesides, G.M. Propulsion of flexible polymer structures in a rotating magnetic field. J. Phys. Condens. Matter 2009, 21, 204110.
  84. Kang, D.J.; An, S.; Yarin, A.L.; Anand, S. Programmable soft robotics based on nano-textured thermo-responsive actuators. Nanoscale 2019, 11, 2065–2070.
  85. Bilodeau, R.A.; Miriyev, A.; Lipson, H.; Kramer-Bottiglio, R. All-soft material system for strong soft actuators. In Proceedings of the 2018 IEEE International Conference on Soft Robotics (RoboSoft), Livorno, Italy, 24–28 April 2018.
  86. Renata, C.; Huang, W.M.; He, L.W.; Yang, J.J. Shape change/memory actuators based on shape memory materials. J. Mech. Sci. Technol. 2017, 31, 4863–4873.
  87. Lobo, P.S.; Almeida, J.; Guerreiro, L. Shape Memory Alloys Behaviour: A Review. Procedia Eng. 2015, 114, 776–783.
  88. Mu, T.; Liu, L.; Lan, X.; Liu, Y.; Leng, J. Shape memory polymers for composites. Compos. Sci. Technol. 2018, 160, 169–198.
  89. Florea, L.; Diamond, D.; Benito-Lopez, F. Photo-Responsive Polymeric Structures Based on Spiropyran. Macromol. Mater. Eng. 2012, 297, 1148–1159.
  90. Verpaalen, R.C.P.; da Cunha, M.P.; Engels, T.A.P.; Debije, M.G.; Schenning, A.P.H.J. Liquid Crystal Networks on Thermoplastics: Reprogrammable Photo-Responsive Actuators. Angew. Chem. Int. Ed. 2020, 59, 4532–4536.
  91. Anam, K.; Al-Jumaily, A.A. Active Exoskeleton Control Systems: State of the Art. Procedia Eng. 2012, 41, 988–994.
  92. Proietti, T.; Crocher, V.; Roby-Brami, A.; Jarrasse, N. Upper-Limb Robotic Exoskeletons for Neurorehabilitation: A Review on Control Strategies. IEEE Rev. Biomed. Eng. 2016, 9, 4–14.
  93. Pons, J. Wearable Robots; Wiley: Hoboken, NJ, USA, 2008; p. 360.
  94. Lo, H.S.; Xie, S.Q. Exoskeleton robots for upper-limb rehabilitation: State of the art and future prospects. Med. Eng. Phys. 2012, 34, 261–268.
  95. Yang, X.; Lihua, G.; Yang, Z.; Gu, W. Lower Extreme Carrying Exoskeleton Robot Adative Control Using Wavelet Neural Networks. In Proceedings of the 2008 Fourth International Conference on Natural Computation, Jinan, China, 18–20 October 2008.
  96. Agarwal, P.; Deshpande, A.D. Impedance and force-field control of the index finger module of a hand exoskeleton for rehabilitation. In Proceedings of the 2015 IEEE International Conference on Rehabilitation Robotics (ICORR), Singapore, 11–14 August 2015.
  97. Cavallaro, E.; Rosen, J.; Perry, J.; Burns, S.; Hannaford, B. Hill-Based Model as a Myoprocessor for a Neural Controlled Powered Exoskeleton Arm—Parameters Optimization. In Proceedings of the 2005 IEEE International Conference on Robotics and Automation, Barcelona, Spain, 18–22 April 2005.
  98. Rosen, J.; Fuchs, M.B.; Arcan, M. Performances of Hill-Type and Neural Network Muscle Models—Toward a Myosignal-Based Exoskeleton. Comput. Biomed. Res. 1999, 32, 415–439.
  99. Rosen, J.; Brand, M.; Fuchs, M.; Arcan, M. A myosignal-based powered exoskeleton system. IEEE Trans. Syst. Man Cybern. Part A Syst. Hum. 2001, 31, 210–222.
  100. Kwon, B.; il Kim, S. Recursive Optimal Finite Impulse Response Filter and Its Application to Adaptive Estimation. Appl. Sci. 2022, 12, 2757.
  101. Shafer, B.A.; Philius, S.A.; Nuckols, R.W.; McCall, J.; Young, A.J.; Sawicki, G.S. Neuromechanics and Energetics of Walking with an Ankle Exoskeleton Using Neuromuscular-Model Based Control: A Parameter Study. Front. Bioeng. Biotechnol. 2021, 9, 615358.
  102. Song, G.; Huang, R.; Qiu, J.; Cheng, H.; Fan, S. Model-based Control with Interaction Predicting for Human-coupled Lower Exoskeleton Systems. J. Intell. Robot. Syst. 2020, 100, 389–400.
  103. Huang, R.; Cheng, H.; Guo, H.; Chen, Q.; Lin, X. Hierarchical Interactive Learning for a HUman-Powered Augmentation Lower EXoskeleton. In Proceedings of the 2016 IEEE International Conference on Robotics and Automation (ICRA), Stockholm, Sweden, 16–21 May 2016.
  104. Dinh, B.K.; Xiloyannis, M.; Antuvan, C.W.; Cappello, L.; Masia, L. Hierarchical Cascade Controller for Assistance Modulation in a Soft Wearable Arm Exoskeleton. IEEE Robot. Autom. Lett. 2017, 2, 1786–1793.
  105. Copaci, D.; Serrano, D.; Moreno, L.; Blanco, D. A High-Level Control Algorithm Based on sEMG Signalling for an Elbow Joint SMA Exoskeleton. Sensors 2018, 18, 2522.
  106. Nef, T.; Guidali, M.; Riener, R. ARMin III—Arm therapy exoskeleton with an ergonomic shoulder actuation. Appl. Bionics Biomech. 2009, 6, 127–142.
  107. Balasubramanian, S.; Wei, R.; He, J. RUPERT closed loop control design. In Proceedings of the 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Vancouver, BC, Canada, 20–24 August 2008.
  108. Tsukahara, A.; Hasegawa, Y.; Sankai, Y. Gait support for complete spinal cord injury patient by synchronized leg-swing with HAL. In Proceedings of the 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems, San Francisco, CA, USA, 25–30 September 2011.
  109. Winkler, A.; Suchý, J. Robot Force/Torque Control in Assembly Tasks. IFAC Proc. Vol. 2013, 46, 796–801.
  110. Hogan, N. Impedance Control: An Approach to Manipulation: Part I—Theory. J. Dyn. Syst. Meas. Control 1985, 107, 1–7.
  111. Unluhisarcikli, O.; Pietrusinski, M.; Weinberg, B.; Bonato, P.; Mavroidis, C. Design and control of a robotic lower extremity exoskeleton for gait rehabilitation. In Proceedings of the 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems, San Francisco, CA, USA, 25–30 September 2011.
  112. Carignan, C.R.; Roderick, S.N.; Naylor, M.P. Distributed Control and Safety System for a Rehabilitation Arm Exoskeleton. In Proceedings of the ASME 2007 International Mechanical Engineering Congress and Exposition, Seattle, DC, USA, 11–15 November 2007.
  113. Wu, Q.; Wang, X.; Du, F.; Xi, R. Modeling and position control of a therapeutic exoskeleton targeting upper extremity rehabilitation. Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci. 2016, 231, 4360–4373.
  114. Carignan, C.; Tang, J.; Roderick, S. Development of an exoskeleton haptic interface for virtual task training. In Proceedings of the 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, St. Louis, MO, USA, 11–15 October 2009.
  115. Frisoli, A.; Borelli, L.; Montagner, A.; Marcheschi, S.; Procopio, C.; Salsedo, F.; Bergamasco, M.; Carboncini, M.C.; Tolaini, M.; Rossi, B. Arm rehabilitation with a robotic exoskeleleton in Virtual Reality. In Proceedings of the 2007 IEEE 10th International Conference on Rehabilitation Robotics, Noordwijk, The Netherlands, 12–15 June 2007.
  116. Nef, T.; Mihelj, M.; Kiefer, G.; Perndl, C.; Muller, R.; Riener, R. ARMin—Exoskeleton for Arm Therapy in Stroke Patients. In Proceedings of the 2007 IEEE 10th International Conference on Rehabilitation Robotics, Noordwijk, The Netherlands, 12–15 June 2007.
  117. Rahman, M.H.; K-Ouimet, T.; Saad, M.; Kenne, J.P.; Archambault, P.S. Tele-operation of a robotic exoskeleton for rehabilitation and passive arm movement assistance. In Proceedings of the 2011 IEEE International Conference on Robotics and Biomimetics, Karon Beach, Thailand, 7–11 December 2011.
  118. Park, J.; Kim, S.J.; Kyeong, S.; Kim, J.; Na, Y. Feasibility of proportional EMG control for a hand exoskeleton: A Fitts’ Law approach. IFAC-PapersOnLine 2018, 51, 214–219.
  119. Xu, D.; Feng, Y.; Mai, J.; Wang, Q. Real-Time On-Board Recognition of Continuous Locomotion Modes for Amputees with Robotic Transtibial Prostheses. IEEE Trans. Neural Syst. Rehabil. Eng. 2018, 26, 2015–2025.
  120. Venkatagiri, R. Optimization of Robotic Exoskeleton with Ameliorated Stability System. Int. J. Innov. Sci. Res. Technol. 2017, 2, 522–525.
  121. Naik, G.R.; Pendharkar, G.; Nguyen, H.T. Wavelet PCA for automatic identification of walking with and without an exoskeleton on a treadmill using pressure and accelerometer sensors. In Proceedings of the 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Orlando, FL, USA, 16–20 August 2016.
  122. Simon, A.M.; Ingraham, K.A.; Spanias, J.A.; Young, A.J.; Finucane, S.B.; Halsne, E.G.; Hargrove, L.J. Delaying Ambulation Mode Transition Decisions Improves Accuracy of a Flexible Control System for Powered Knee-Ankle Prosthesis. IEEE Trans. Neural Syst. Rehabil. Eng. 2017, 25, 1164–1171.
  123. Zhang, K.; de Silva, C.W.; Fu, C. Sensor Fusion for Predictive Control of Human-Prosthesis-Environment Dynamics in Assistive Walking: A Survey. arXiv 2019.
  124. Pratt, J.; Krupp, B.; Morse, C.; Collins, S. The RoboKnee: An exoskeleton for enhancing strength and endurance during walking. In Proceedings of the IEEE International Conference on Robotics and Automation, ICRA ’04, New Orleans, LA, USA, 26 April–1 May 2004.
  125. Kong, K.; Jeon, D. Design and control of an exoskeleton for the elderly and patients. IEEE/ASME Trans. Mechatron. 2006, 11, 428–432.
  126. Fleischer, C.; Hommel, G. A Human–Exoskeleton Interface Utilizing Electromyography. IEEE Trans. Robot. 2008, 24, 872–882.
  127. Peternel, L.; Noda, T.; Petrič, T.; Ude, A.; Morimoto, J.; Babič, J. Adaptive Control of Exoskeleton Robots for Periodic Assistive Behaviours Based on EMG Feedback Minimisation. PLoS ONE 2016, 11, e0148942.
  128. Rettinger, J.; Schwarz, S.; Schwarz, W. Electrophysiology Basics, Modern Approaches and Applications; Springer: Cham, Switzerland, 2016; p. 175.
  129. André, A.D.; Teixeira, A.M.; Martins, P. EMG Signals as a Way to Control Soft Actuators. In Computer Methods, Imaging and Visualization in Biomechanics and Biomedical Engineering II; Springer International Publishing: Cham, Switzerland, 2022; pp. 50–58.
  130. Péter, A.; Arndt, A.; Hegyi, A.; Finni, T.; Andersson, E.; Alkjær, T.; Tarassova, O.; Rönquist, G.; Cronin, N. Effect of footwear on intramuscular EMG activity of plantar flexor muscles in walking. J. Electromyogr. Kinesiol. 2020, 55, 102474.
  131. Qi, J.; Jiang, G.; Li, G.; Sun, Y.; Tao, B. Surface EMG hand gesture recognition system based on PCA and GRNN. Neural Comput. Appl. 2019, 32, 6343–6351.
  132. Oskoei, M.A.; Hu, H. Myoelectric control systems—A survey. Biomed. Signal Process. Control 2007, 2, 275–294.
  133. Lenzi, T.; Rossi, S.M.M.D.; Vitiello, N.; Carrozza, M.C. Intention-Based EMG Control for Powered Exoskeletons. IEEE Trans. Biomed. Eng. 2012, 59, 2180–2190.
  134. Noda, T.; Furukawa, J.; Teramae, T.; Hyon, S.H.; Morimoto, J. An electromyogram based force control coordinated in assistive interaction. In Proceedings of the 2013 IEEE International Conference on Robotics and Automation, Karlsruhe, Germany, 6–10 May 2013.
  135. Farina, D.; Merletti, R.; Indino, B.; Nazzaro, M.; Pozzo, M. Surface EMG crosstalk between knee extensor muscles: Experimental and model results. Muscle Nerve 2002, 26, 681–695.
  136. Wilson, S.; Vaidyanathan, R. Upper-limb prosthetic control using wearable multichannel mechanomyography. In Proceedings of the 2017 International Conference on Rehabilitation Robotics (ICORR), London, UK, 17–20 July 2017.
  137. Orizio, C. Muscle sound: Bases for the introduction of a mechanomyographic signal in muscle studies. Crit. Rev. Biomed. Eng. 1993, 21, 201–243.
  138. Dallali, H.; Demircan, E.; Rastgaar, M. Powered Prostheses Design, Control, and Clinical Applications; Elsevier Science and Technology: Amsterdam, The Netherlands, 2020.
  139. Rabe, K.G.; Jahanandish, M.H.; Hoyt, K.; Fey, N.P. Use of Sonomyography for Continuous Estimation of Hip, Knee and Ankle Moments During Multiple Ambulation Tasks. In Proceedings of the 2020 8th IEEE RAS/EMBS International Conference for Biomedical Robotics and Biomechatronics (BioRob), New York, NY, USA, 29 November–1 December 2020.
  140. He, J.; Luo, H.; Jia, J.; Yeow, J.T.W.; Jiang, N. Wrist and Finger Gesture Recognition with Single-Element Ultrasound Signals: A Comparison with Single-Channel Surface Electromyogram. IEEE Trans. Biomed. Eng. 2019, 66, 1277–1284.
  141. Noda, T.; Sugimoto, N.; Furukawa, J.; Sato, M.A.; Hyon, S.H.; Morimoto, J. Brain-controlled exoskeleton robot for BMI rehabilitation. In Proceedings of the 2012 12th IEEE-RAS International Conference on Humanoid Robots (Humanoids 2012), Osaka, Japan, 29 November–1 December 2012.
More
© Text is available under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/)
Video Production Service
Feedback