Next Article in Journal
Determination of Arsenic Species Distribution in Arsenide Tailings and Leakage Using Geochemical and Geophysical Methods
Previous Article in Journal
Evaluation of the Effect of Stearyl Alcohol and Span-60 Tuned Sunflower Wax/Sunflower Oil Oleogel on Butter Replacement in Whole Wheat Cake
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 13
Issue 2
10.3390/app13021064
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Review

Mechanical Structural Design and Actuation Technologies of Powered Knee Exoskeletons: A Review

by
Zongpeng Wu
1,2,3,
Mingxing Yang
1,2,4,5,*,
Yulei Xia
1,4 and
Liwei Wang
1,4
1
Anhui Province Key Laboratory of Special Heavy Load Robot, Anhui University of Technology, Maanshan 243032, China
2
Anhui Province Engineering Laboratory of Intelligent Demolition Equipment, Anhui University of Technology, Maanshan 243032, China
3
Engineering Practice and Innovation Education Center, Anhui University of Technology, Maanshan 243032, China
4
School of Mechanical Engineering, Anhui University of Technology, Maanshan 243032, China
5
Haisida Robot Company of Anhui Province, Maanshan 243000, China
*
Author to whom correspondence should be addressed.
Appl. Sci. 2023, 13(2), 1064; https://doi.org/10.3390/app13021064
Submission received: 21 December 2022 / Revised: 6 January 2023 / Accepted: 10 January 2023 / Published: 12 January 2023
Browse Figures
Versions Notes

Abstract

:
Robot knee exoskeletons can not only help the rehabilitation training function of the elderly and disabled patients, but also enhance the performance of healthy people in normal walking and weigh-bearing walking by providing sufficient torques. In recent years, the exoskeletons of knee joints have been extensively explored. The review is to summarize the existing research results of mechanical structure design and actuation technologies, propose the future development trend, and promote the further development of the powered knee exoskeletons, related theories, and engineering applications. In this study, the mechanical structures of knee exoskeletons are first illustrated. Their mechanical structures are classified into two types: simple mechanical structures with one purely rotary DOF and biological geometry-based multi-DOF structures. Subsequently, the actuation design of wearable knee exoskeletons includes conventional driving actuators, pneumatic muscle actuators, variable stiffness actuators, and other actuators are compared and the driving compliance and the difficulty in the accurate control are analyzed. Furthermore, other crucial technologies such as motion intention recognition, control strategy and performance evaluation methods of most knee assistive devices are reviewed. Finally, the key technologies of structural design and actuation design in the research of knee exoskeletons are summarized and future research hotspots are proposed.

1. Introduction

Nowadays, the development of medical technologies and the improvements of living conditions bring convenience and improves quality of life, but the situation of population aging is not optimistic. In the aging process, the gradual decline of lower limb motor function and the mobility inconvenience caused by other diseases have largely affected the life of the elderly [1,2]. In addition, the contradiction between the increasing number of patients with limb injuries caused by stroke and traffic accidents and the current shortage of medical resources (the gap of professional medical nursing personnel has reached millions) is increasingly prominent. Exoskeleton robots, which combine human intelligence with robot power, are an important technical path to eliminate these challenges and alleviate the current medical dilemma. The lower limb exoskeleton robot forms an intelligent device for man-machine coordination by parallel connection with the wearer and utilizes the human action guidance function and specific programs to assist the wearer in walking or rehabilitation training [3,4,5]. Therefore, as an intelligent man-machine integration system, the lower limb exoskeleton robot is significant in the fields of helping the disable and the elderly, medical rehabilitation, and industrial production.
Various exoskeleton robots have been developed according to functional requirements in different fields, including developing different types of joint orthotics and exoskeletons to aid the movement of ankle, knee, and hip joint [6,7,8]. In particular, as the most vulnerable joint of the human body, the knee joint not only supports walking, running, and squatting to standing, but also absorbs the impact of human walking [9]. It is noteworthy that the dysfunction of a knee joint caused by diseases and injuries is the most common factor for gait abnormalities and seriously affects patients’ living ability and mental health [10]. Therefore, many methods have been proposed to reduce the internal force of the knee joint, carry out rehabilitation exercises after the knee joint injury, and develop the corresponding exoskeleton for improving the exercise ability.
In the past few decades, many universities and scientific institutions have developed active and semi-active knee joint exoskeletons to help users recover their walking ability and strengthen their muscles in a more natural way [11,12,13,14,15,16,17,18,19,20]. According to the intended use, the powered knee joint exoskeleton robots can be classified into exoskeletons for walking aids for the wear and therapeutic exoskeletons for rehabilitation. The first type of knee exoskeleton is mainly designed for patients with stroke, spinal cord injury, muscle weakness or other diseases that may lead to walking difficulties because they allow patients to complete some actions that they cannot implement independently [11,12,13,15]. The second type of knee exoskeleton can help, resist or interfere with the user’s movement to achieve therapeutic exercise and train the individual’s muscle and nervous system to overcome the limitations of disability [14,16,17,18]. From another point of view, some knee exoskeleton devices have the functions of both assisted walking and therapeutic rehabilitation, and can help medical treatment and improve the current physical ability of the wearer [19,20].
However, the application scope of these devices is limited by several factors. In particular, wearing comfort, energy supply, actuator design, and adaptive compliance control of all exoskeletons need to be improved [21,22,23]. The review summarizes the design and development of mechanical structures and drive actuators related to the knee joint exoskeletons for rehabilitation and enhancement of human performance, and points out possible research directions to improve existing devices. In addition, the challenges in the application of wearable knee exoskeleton in various fields (such as rehabilitation, medical care, and disabled people) are introduced, and relevant solutions and methods are also proposed.
The rest of this manuscript is organized as follows. In Section 2, the mechanical structural design of powered knee exoskeletons is described in detail based on biological geometries of the knee joint. According to the power transmission design and structural characteristics, the actuation design of knee assistive devices is introduced in Section 3. Section 4 provides the descriptions of other key technologies, motion intention recognition, control strategies, and performance evaluation. In Section 5, the significant findings of knee joint exoskeleton robots in the structural design and actuation technologies are discussed and the future research hotspots are presented.

2. Mechanical Structural Design of Knee Exoskeletons

As shown in Figure 1a, from the anatomical standpoint, a human knee joint mainly consists of tibia, patella, femur, cartilage, surrounding ligaments, and muscles. In addition, femur and tibia at the knee joint are connected by ligaments and their sophisticated movements can lead to sliding and rocking motions as well as rotation motion. Compared with hip joint and ankle joint, a knee joint, as the part connecting the thigh and lower leg, has more direct flexion/extension and rotation movements. According to the physiological characteristics and biomechanical structure analysis of the human body, the geometric structure of a knee joint is uneven with varying articulating surfaces, showing a multi-center motion with a non-constant rotation axis in the sagittal plane [24]. In particular, when a knee joint bends, the femur slides and rolls on the tibia with a variable instantaneous rotation center on the sagittal plane [25], as illustrated in Figure 1b. Therefore, the kinematics of biological joints should be fully considered before designing a bionic exoskeleton knee joint structure, otherwise the design of a lower-extremity exoskeleton with insufficient knee-motion knowledge may disturb and even damage the human knee.

2.1. Simplified Structure Design

Although the knee is a complex joint, most degree of freedom (DOF) of the knee joint can be ignored and only the flexion/extension movement in the sagittal plane is considered in biomechanical studies [26,27]. Hence, to simplify the mechanical structure and promote the system control by simplifying the dynamic model, a single DOF motion has been designed for the knee exoskeleton based on the assumption that simplify bio-joints to engineering joints. In general, a simple single DOF hinge structure is mainly used to assist the flexion/extension motion of the knee joint in the sagittal plane of the human body.
For example, as a portable mechanism for knee exercises, the ERF device utilized a simple revolute model driven by an electro-rheological fluid-based rotary actuator at the knee joint [28], as shown in Figure 2a. The energetically autonomous powered RoboKnee exoskeleton [29], a wearable device specific designed to facilitate running, was constructed with a motor-driven screw bars-based rotary actuator, which also adopted a simple hinge model for the knee joint, as shown in Figure 2b. On the basis of the pure rotary DOF modeling of human knee, EICOSI orthosis employed two cables with antagonistic actuation to provide sufficient torque for the knee joint [30], as described in Figure 2c. In other work, a compliant support exoskeleton with elastic energy storage called KAFO was developed in [31], and the feasibility of compliance assistance of the designed 1-DOF knee joint was experimentally verified based on quasi-passive compliant stance control, as shown in Figure 2d. A powered exoskeleton using Inertial Measurement Unit (IMU) sensors was developed to provide effective assistance during walking for impaired users with knee dysfunction in [32], as shown in Figure 2e, the active knee joint of the mechanical system was designed as a perfect revolute joint actuated by a high torque motor and a low-ratio gear transmission. Similarly, in [33] a new type of knee exoskeleton consisting of a thigh segment and a shank segment was designed and its unique DOF of rotation was adjusted by a hydraulic damper. In order to provide walking assistance and reducing the metabolic demand of locomotion, a parallel-elastic knee exoskeleton operated by a custom interference clutch with integrated planetary gear transmission was proposed in [34].
Additionally, in some commercial exoskeletons, one purely rotary motion was also arranged at the knee joint, such as the Reo Ambulator used for enhancing strength and endurance during walking [35], the most well-known automated gait trainers Lokomat [36], and another popular gait rehabilitation robot LOPES [37]. It is worth noting that although the exoskeleton or orthosis mentioned above can help or adjust the human musculoskeletal system, it may cause discomfort and injury if it is not adaptive to the wearer.

2.2. Complex Design Based on Biological Geometries

In fact, the human knee joint, as a complex multi-center form, contributes to its highly flexible and adaptable human lower limb locomotion. However, the single mechanical rotation model for the knee joint is prone to cause joint dislocation when the rotation centers of human body and the machine do not coincide with each other [38]. In the last decade, a number of gait rehabilitation devices were developed to eliminate negative effects of the closed leg-exoskeleton kinematic chain on a human knee.
The multi-bar mechanism can provide a polycentric motion similar to the human knee joint to achieve self-alignment, so it can reduce the potential discomfort caused by the movement inconsistency between the wearer and assistive devices. Kim J H et al. [11] proposed a modular knee exoskeleton system to support the knee joint of hemiplegic patients, as shown in Figure 3a. This device aims to realize the multi-center movement of the real human knee joint through a four-bar mechanism. For the sake of providing redundant DOFs to compensate for the knee joint, Saccares L et al. [39] explored a power-aided exoskeleton for the knee joint utilizing two generalized conforming quadrilateral structures. To well adjust the translational motion of the knee joint and allow for a natural gait, Eschbach M et al. designed a novel mechanism that combined a four-bar linkage and a prismatic joint into a two-degree-of-freedom hinge for knee osteoarthritis [40]. However, increasing the degree of freedom by increasing connecting rods inevitably led to the complexity and inertia of the structural design.
Meanwhile, Choi B et al. [41] presented a Gait-Enhancing Mechatronic System (GEMS) equipped with a pulley and several rolling cams to adapt to the joint motion of a wearer, as shown in Figure 3b, the articulated joint could effectively deliver the required torque and prevent discomfort in the process of physical gait assistance. Based on the knowledge of a knee-joint kinematics, Wang D et al. [9] introduced an adaptive knee-joint exoskeleton that incorporated a pin slider/cam, and the investigation indicated that it could effectively minimize internal joint forces and torque via the human–machine interaction. Besides, to achieve the alignment with the instantaneous knee center of rotation, Terada H [42] and Younbaek Lee et al. [43] developed a wearable walking assistance device employing cam and slider mechanisms. These optimized mechanisms were designed to replicate the reference path of the human knee, however, the misalignment is the main problem to be considered that other design objectives have not been considered.
Pulley, belt, and other mechanical parts are combined together to enhance the flexibility and self-aligning ability of the knee joint. To provide effective torque assistance for the knee joint without disrupting the natural gait of the wear, Wang J et al. [44] developed a personal joint mechanism in conjunction with two-stage timing belt transmission system, as shown in Figure 3c. Ergin M A et al. [45] introduced a novel knee exoskeleton with three co-centered rings driven through a belt drive transmission for robot-assisted rehabilitation and indicated that the robot was capable of accommodating transitional movements of the knee joint along with its rotation. Obviously, the intelligent design utilizing the belt and machine parts can also effectively minimize or eliminate the misalignment between robot and human knee joint. However, these knee assistance devices are not ideal due to the complexity and large size of the belt transmission system.
Furthermore, many other mechanisms have been designed to adapt to the mismatching between the instantaneous center of the rotation of human lower limb locomotion and the knee exoskeletons. As shown in Figure 3d, Celebi B et al. [46] developed a self-aligned active exoskeleton with an under-actuated Schmidt coupling for robot-assisted knee rehabilitation. The Schmidt coupling is composed of seven rigid bodies, which is a planar mechanism with three degrees of freedom (ie., one active rotary motion dominated by a Bowden cable driven series elastic actuator, and the other two passive translational motions in the sagittal plane). In another work, Kim T et al. [47] proposed a bioinspired knee joint with a curved guide rail and three ligament bearing for rehabilitation, as shown in Figure 3e, the robot is capable of closely realizing the human knee joint motion and prevent injury and discomfort during active assistance. Moon DH et al. [48] designed a modular knee exoskeleton with the lower frame length, which could be adjusted to be adapted to the complex motion of the knee so as to reduce the weight of the driving section directly attached on the knee. Li HW et al. [49] designed a self-adapting compliant joint with a rolling pair to generate an ideal instantaneous center of rotation trajectory, thereby reducing the joint misalignment for the gait enhancing mechatronics system.
Table 1 give an overview of the simplified structure design and the complex design based on biological geometries of existing knee exoskeletons. Both the simplified structure of the 1-DOF knee exoskeleton or that structure based on biological geometries have typical applications in many fields. Therefore, an appropriate exoskeleton of the knee joint should be selected according to the specific requirements for structural design. In the field of rehabilitation training for the patients with knee joint dysfunction, patients and exoskeletons are strictly bound together, so the rotation center of the device should be strictly consistent with that of the joint center of a wearer. Similarly, when knee exoskeletons are used to assist walking or running for healthy people, it is necessary to focus on reducing the inertia and weight of the overall structure as possible without affecting comfort. Therefore, many existing exoskeleton robots ignore this coupling effect and adopt a simple hinge with a pure rotary DOF to mimic the knee joint.

3. Actuation Design

The driving mode directly affects the structural design, control method, and other system schemes of the power-assisted exoskeleton robots and is important in the system design. As a human-computer interaction device, an exoskeleton actuator should have sufficient torque output, reliability, mechanical compliance, and even adaptability toward the wearer [50]. At present, the studies on driving modes of knee exoskeletons mainly focused on motor cable drive, the gradually mature bionic pneumatic muscle drive, and the new actuators based on diverse flexible materials [51,52,53,54]. According to their power transmission design and structural characteristics, the actuators of knee joint assistance devices can be divided into four categories: conventional driving actuators (CDAs), pneumatic muscle actuators (PMAs), variable stiffness actuators (VSAs), and other actuators.

3.1. Conventional Driving Actuators

The conventional driving models of powered knee exoskeletons are mainly linear hydraulic actuators, pneumatic cylinder and electric actuators.
As shown in Figure 4a, Zhu J et al. [55] designed a unidirectional variable stiffness hydraulic actuator for load-carrying knee exoskeleton, which can mimic the efficient passive behavior of the knee joint and provide active assistance in locomotion. In order to provide mobility and independence to the elderly and the disabled, Kaminaga H et al. [56] developed a prototype of power-assisted knee joint exoskeleton. Its man-machine conversation system was equipped with a hydraulic actuator specially designed to realize backdrivability and improve the discomfort caused by the errors between human and robot motion. Besides, Klein Rot J E et al. [57] introduced an active knee exoskeleton utilizing a pneumatic cylinder for the purpose of assisting the wearer while walking, as shown in Figure 4b, the experimental results show that the assistant strategy performed well in gait interference. To provide power assistance for leg swing and prevents knee buckling during stance, Aoyagi D et al. [58] proposed a pneumatically operated gait orthosis which was actuated by pneumatic cylinders. Nevertheless, according to the working principle of hydraulic cylinder or pneumatic cylinder, the fluid with a certain pressure was applied on the piston rod to generate strong power output, but it required the device with a large power and had a low efficiency [59]. Therefore, these types of assistive wearing devices mainly depended on sufficient power to support the wearer with great assistance, but they had obvious defects in the overall weight and drive flexibility.
The electric actuator, as another conventional driving model, has the advantage of controllability that can be used for the precise movement of exoskeletons. Therefore, the exoskeleton of a knee joint can be powered by a motor in combination with other transmission mechanisms, such as gears, ball screws, Bowden cables, and belts [60,61]. In order to provide walking assistance for people with knee joint disorders, Jain P et al. [62] designed an electromyography sensor-based knee exoskeleton actuated by a linear actuator, as shown in Figure 4c. In the bond graph model of fixed causality, the actuator with a linear DC motor lies in the electrical domain, whereas the gearbox, lead screw, and four-bar mechanism stay in the mechanical domain. To seek out an ideal actuator with a high torque, a high efficiency, and a small size for the knee exoskeleton, Horst R and Marcus R [63] developed a new type of actuator with direct current (DC) motors and batteries, which could provide continuously varying torque through two belts alternately deflected by cams. Kong K et al. [64] proposed a compact and wearable exoskeleton with a servo motor-driven module actuator to generate an assistive torque for human knee joint, as shown in Figure 4d, the actuator was a compact series elastic device that utilized a torsional spring in the chain of spur gears and worm gears. Chen G et al. [65] introduced a novel compact series elastic actuator installed at the knee joint for stroke patients to perform gait training. As shown in Figure 4e, this actuator was employed a servo motor to drive ball screw and two translational springs with different stiffness are placed in series for force transmission to provide assistance during flexion-extension movements of the knee. More recently, Zhou C et al. [66] designed a knee exoskeleton with a series elastic actuator, which is capable of providing variable joint stiffness in response to changing external demands. Note that the compliant joint of exoskeletons with series elastic actuator could realize the performance of low impedance, however, the fixed stiffness of the actuator limited its agility performance.

3.2. Pneumatic Muscle Actuators

Pneumatic muscle actuators inflate a sheathed bladder by employing a compressed gas to make it contract lengthwise when expanded radially and have been successfully applied in robotics and robotic therapy. In comparison with conventional driving actuators in human-robot interaction, the pneumatic muscle actuator has unique advantages including light weight, simple structure and intrinsic compliance [67]. Thus, the application of pneumatic artificial muscle in the knee exoskeleton robots have been explored and a large number of related prototypes have been developed.
Beyl P et al. [68] developed a unilateral exoskeleton with a knee joint to assist the rehabilitation of gait, as shown in Figure 5a. The powered exoskeleton was actuated by two antagonistically pleated pneumatic artificial muscles, and the force was transmitted to the lower leg link by means of two four-bar linkages. In order to realize a wearable robot for enhancing human walking, Maeda D et al. [69] developed a robotic knee exoskeleton with soft McKibben style pneumatic artificial muscles (PAMs). As shown in Figure 5b, the actuator could improve the compliance of the exoskeleton by controlling the internal air pressures of agonist-antagonist PAMs and respectively adjusting its balance point and stiffness linearly. In addition, Mohri S et al. [70] developed a powered suit actuated by straight-fiber-type artificial muscles for knee auxiliary, as shown in Figure 5c. The assistive suit was composed of wire, pneumatic artificial muscles, and some fixed parts, and it could generate torque in the knee joint as the rotation center in case the wire was tensioned by the contraction of the artificial muscle. As shown in Figure 5d, Veale A J et al. [71] developed a flexible wearable exoskeleton with a novel pleated pneumatic interference actuator to assist knee movement. As a soft structure, the actuator could stably and comfortably allow the knee to generate sufficient forces for extension and flexion. Cao J et al. [72] developed a compliant exoskeleton system which utilized artificial pneumatic muscles to provide torque assistance at the knee, as shown in Figure 5e, and the experiment showed that the actuator performed more positive mechanical work than negative mechanical work compared to normal walking conditions without the device. Besides, many other exoskeletons have been designed to provide torque assistance to the knee joint through pneumatic muscle actuators, such as a lower-limb exoskeleton with hybrid actuation for the strength augmentation [73], a power exoskeleton to assist weakened individuals during locomotor training [74], and an assistive exoskeleton system [75] for normal walking.
As mentioned above, pneumatic muscle actuators are suitable for the exoskeleton type of robotic devices due to their intrinsic compliance and large force-to-weight ratio. Nevertheless, pneumatic muscles-based system may not be further developed because the pneumatic power supply device could bring a burden and generate noise. In addition, considering that the dynamics of PMAs is highly nonlinear and subjected to hysteresis, these developed knee exoskeletons may produce uncomfortable wearing experiences due to the difficulty in precise tracking control.

3.3. Variable Stiffness Actuators

In the applications of exoskeletons, variable stiffness actuators (VSAs) as another optional driving model allows the adaptation to environmental changes and task requirements through changing the output stiffness of the active joint [76]. Compared with other types of actuators, variable stiffness actuators can more efficiently imitate the nonlinear characteristics of the human joints to improve the performance of shock absorption and reduce energy consumption. Therefore, numerous knee exoskeletons with compliant joints exhibiting variable stiffness performances have recently been proposed.
Shamaei K et al. [77] designed a knee exoskeleton integrating flexible drive with wearable mechanism to help users recover their walking ability. As shown in Figure 6a, the variable-stiffness module equipped with two springs mounted on an adjustable knee brace allowed the wearer’s knee joint to move freely in the terminal stance and the swing phase. As shown in Figure 6b, Bacek T et al. [78] proposed a novel compliant actuator that powers the knee joint by employing springs working both in series and in parallel to the joint. Note that the parallel elastic element was a quasi-passive mechanism composed of an EC motor, a spindle, and a spring, whereas the series elastic element was a variable stiffness actuator which remained engaged throughout the gait cycle. To effectively assist people in walking or perform rehabilitation training, Tian M et al. [79] designed a wearable exoskeleton actuated by tendon-sheath artificial muscles, as shown in Figure 6c. The actuator completed the contraction and relaxation of the artificial muscle with the controllable clutch to adjust series and parallel springs, which was utilized to imitate the output characteristics of the muscle. For the purpose of helping patients with knee motor dysfunction recover their walking ability, Chen B et al. [80] designed a robotic knee exoskeleton with a variable stiffness actuator, as shown in Figure 6d. As the elastic element of the actuator, a disc-type torsion spring was designed in the compliance module, thus enabling the mechanical device to reduce the output impedance of the system. Furthermore, Jafari A et al. [76] proposed a novel intrinsic energy-efficient actuator to achieve the stiffness regulation by using the variable lever arm principle. To minimize the large contact force and safely interact with the user, Cestari M et al. [81] conceived a new force-controlled compliant actuator with an embedded sensor and a locking mechanism. Regarding the driving compliance of the knee joint exoskeleton, Baček T et al. [82] analyzed the influence of the mechanical flexibility on the performance of the actuator under different ideal conditions in the experimental setup.
In terms of safety and compliance of human-robot interaction, the proposed variable stiffness actuators can effectively regulate the stiffness of the knee exoskeleton according to the requirements under most conditions. This is possible due to their novel mechanical configurations make full use of natural dynamics to achieve the stiffness regulation and deliver biologically-relevant knee joint torque-angle output. However, to the best of our knowledge, the ability of VSA-based actuation units to realize driving characteristics found in biological systems usually comes at the cost of higher weight, complexity, and lack of robustness.

3.4. Other Actuators

In recent studies, smart and composite materials have demonstrated enormous potential in knee exoskeletons to realize compliant human-robot interaction, including shape memory alloy (SMA) actuators [83], vacuum-actuated rotary actuators [84], and plasticized polyvinyl chloride (PVC) gel soft actuators [85]. These innovations have their own unique means to efficiently imitate the nonlinear characteristics of the human knee joints. Nevertheless, to the best of our knowledge, some essential problems such as nonlinear hysteresis effect, high technical difficulty, and complex control remain to be solved before their further applications.
Based on the above studies, Table 2 shows the comparison results of the actuation designs of existing powered knee exoskeletons, including the characteristics, advantages, and disadvantages of each category. Regardless of the application domain or body area, the actuator should be selected according to the intention or purpose of the designer because each device has its own advantages and disadvantages. In practice, human-computer interaction security, motion flexibility, and adaptability to complex environments are required in wearable knee exoskeletons. Besides, in the design of knee exoskeletons, the tradeoff between weight and the output torque should be fully considered so as to develop a more compact and lightweight mechanism with improved wearing comfort.

4. Other Key Technologies

The main characteristic of the knee exoskeleton is the combination of human intelligence and physical strength and it is an important rehabilitation and assistive tool for patients with walking dysfunction of the knee joint. However, apart from improving the physical interaction between the exoskeleton system and the human body through the bionic compliant structure design and the high-performance actuator technology, other key technologies (motion intention recognition, control strategies, and effective evaluation) should be adopted in the research field of knee exoskeleton robots.

4.1. Motion Intention Recognition

The knee exoskeleton robots and the wearer should achieve precise man-machine cooperation, so the mechanical device can realize the function of strength enhancement or motion assistance as required. The effect of motion intention prediction directly affects the quality of human-robot cooperative control. The core of motion intention recognition is to accurately convey the human motion intention and extract and fuse the feature information as the real-time input signal of the exoskeleton system [23,86].
The gait trajectory of human knee can be estimated with the signals of physical sensors [20,42,71]. The human-robot interaction force is measured with force/displacement sensors and then the useful signals can be utilized as feedback information in the algorithm control of knee exoskeleton through information conversion. This motion intention recognition method can be conveniently implemented, but it is difficult for the exoskeleton and the wearer to work together synchronously due to a large delay. In addition, the inertial measurement unit (IMU) sensors can be used to estimate the upcoming gait trajectory of the knee joint by detecting the body posture information and motion information of the wearer [32,34,35]. Unfortunately, this online knee gait trajectory generation method has the limitation that its initial motion assistance is from the user. Another important way is to use the multi-sensor fusion measurement technology to identify human motion intention [9,31]. In this technology, the posture detection module, acceleration measurement module, and foot pressure detection module are combined together to realize the tracking of human motion with the knee exoskeleton, thus increasing the quantity of information to be processed and reducing the usability of the device. The predicted signals can be detected before the human lower limb produced movement and help the exoskeleton strictly follow the human knee joint, so the bioelectrical signal method based on the electromyography (EMG) signal or electroencephalogram (EEG) signal may be the most potential method for motion intention recognition [6,87,88]. Although the bioelectrical signals have been applied widely in human-robot interfaces, the process of model training and calibration is cumbersome and difficult. Therefore, it is urgent to develop a simple and reliable online motion intention recognition method without various sensors in the further development of the knee exoskeleton.

4.2. Control Strategies

The control system design of the power-assisted wearable knee joint exoskeleton robot is widely concerned because it largely affects the power-assisted effect. At present, a common control strategy of the knee exoskeleton is not available due to its vast diversity in design purposes, hardware structures, and actuation modes [8,89]. In particular, according to the integration idea of the control system, the control strategy of the knee exoskeleton mainly includes offline trajectory setting, online trajectory regulation, and real-time trajectory tracking [90].
The offline trajectory setting is adopted in the medical exoskeleton of knee joint for rehabilitation training of patients and can be completed by manually setting the desired position information or with the gait data of healthy subjects [68,91]. This control strategy is simple and direct, but its efficiency is low and other auxiliary tools are generally used to ensure the balance of the system. Based on the measurements of sensors, the online trajectory regulation of the knee exoskeleton can be executed with model-based control methods [21,92]. This control strategy is usually required to ensure that the trajectory is in an acceptable safe range and that information input is used to identify human intentions for the adaptation to the parameter changes of the real walking scenario. However, the knee exoskeleton is a typical human-machine interaction system, so different users and walking conditions significantly affect its performance. It is necessary to design a real-time trajectory tracking method for target knee exoskeletons based on the measurements of human-machine coupled interaction and motion information of the wearer [30,32,93]. A variety of controllers can be chosen. PID control is a common choice and other choices include adaptive control, sliding mode control, and impedance controller with disturbance rejection. The controller needs to respond to the movement information of the wearer in a timely manner like the human central nervous system, so the control system requires a fast response to the movement of the wearer and minimize human-machine interference [94]. In other words, the knee joint exoskeleton robot and the wearer are interrelated and unified. It is necessary to consider both the driving effect of exoskeleton robot on human body and the leading role of human motion intention, so as to achieve high assistance performance and wear comfort.

4.3. Performance Evaluation

The exoskeleton of the knee mainly provides power to enable the knee joint of humans to move in the sagittal plane, and its working performance determines the feasibility and availability of the power-assisted device. A unified and complete evaluation system of the power-assisted knee joint exoskeleton is not available. In the aspect of performance evaluation, both researchers and engineers should focus on the accurate control and reliability of knee exoskeleton robots.
In general, a simple and direct method is to collect biomechanics information of the knee joint, including necessary dynamic foot-pressure, man-machine contact force, joint turning angle, and other information [44,95,96]. In terms of the specific assistance conditions of a knee exoskeleton robot, the joint motion, moment, and power profiles may exhibit a subtle trend of change. Besides, the most efficient means to evaluate the assistive performance is to detect biological parameters of the wearer, such as heart rate, muscle activity, and oxygen consumption [7,27,97]. In particular, studies have shown that the metabolic reduction is biomechanically caused by the significant reduction in the kinetic effort and extensor muscle activation of the knee joint [98,99,100]. It should be noted that even under different assistance installation conditions, the trends of EMG, biomechanics, or energetics probably showed no significant differences. This indicates that the controllability of the wearer’s assistance profile at the knee joint was not strictly consistent with the purpose of human augmentation during steady-state walking under different conditions [94,101]. At present, metabolic cost, muscle activity, motion capture data, and ground reaction force have been widely considered in the performance evaluation of knee-assisted devices [102]. The evaluation methods discussed above can reflect the partial performance parameters of the exoskeleton, but they ignored the impacts of related test requirements on the wearer’s mental state during the test. In addition, the safety and reliability evaluation of the exoskeleton robot remains to be further explored.

5. Conclusions

Due to the development of computer technology and robotics, the applications of exoskeleton robots in military, medical, and civil fields will become more and more prominent in the future. The paper summarizes recent progress in the structural design and actuation technologies of knee joint exoskeleton robots. The conclusions are drawn as follows:
Firstly, in terms of the structural design of knee exoskeletons, both the simple mechanical structure with one purely rotary DOF and the multi-DOF structure based on biological geometries can be utilized to installed at the moving joint. It is particularly difficult for the exoskeleton to imitate the knee joint motion due to its complex structure, the misalignment issue cannot be avoided with a single revolute joint. With a multi-bar mechanism, a passive pulley or multiple-rolling-cam structure imitating the motion of human knee joint may be used to further improve the bionic performance, but its weight or size might not be suitable for a wearable system.
Secondly, the actuator design and performance evaluation of wearable knee exoskeletons is to realize the satisfying human-robot interaction performance of the intelligent device and improve or enhance the function of human knee joint movement. Therefore, the knee exoskeleton system should have the ideal bionic human-in-the-loop control performance and acquire the comprehensive evaluation information of human motion functions such as heartbeat, muscle hardness, oxygen consumption, and blood pressure. However, the compliance design of actuators is largely realized at the cost of accurate position/force control, and appropriate exoskeleton actuators should be chosen as required.
Thirdly, in the future, the development tendency of knee exoskeletons is the lightweight and drive compliance under the premise of anthropomorphic bionic structure design. Based on the theory of motion intention recognition, multi-mode sensory feedback control strategy and reliability performance evaluation method, a remarkable power-assisted exoskeleton device for humans will be developed.

Author Contributions

Conceptualization, Z.W., M.Y, Y.X. and L.W.; methodology, Z.W. and M.Y.; formal analysis, Z.W. and M.Y.; data curation, Z.W. and M.Y.; writing—original draft preparation, Z.W., Y.X., L.W. and M.Y.; writing—review and editing, Z.W. and M.Y. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Nation Nature Science Foundation of China (grant number 52005006), the Open Project of Anhui Province Engineering Laboratory of Intelligent Demolition Equipment (grant number APELIDE2021B002), and the Postdoctoral Research Fund of Anhui Province (grant number 2021B504).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Acknowledgments

We are grateful to the Nation Nature Science Foundation of China under grant No. 52005006, the Open Project of Anhui Province Engineering Laboratory of Intelligent Demolition Equipment under grant No. APELIDE2021B002, and the Postdoctoral Research Fund of Anhui Province under grant No. 2021B504.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Chen, G.; Chan, C.K.; Guo, Z.; Yu, H. A review of lower extremity assistive robotic exoskeletons in rehabilitation therapy. Critical Review. Biomed. Eng. 2013, 41, 343–363. [Google Scholar]
  2. Chen, W.; Li, J.; Zhu, S.; Zhang, X.; Men, Y.; Wu, H. Gait recognition for lower limb exoskeletons based on interactive information fusion. Appl. Bionics Biomech. 2022, 9933018, 1–19. [Google Scholar] [CrossRef] [PubMed]
  3. Attias, M.; Bonnefoy-Mazure, A.; De Coulon, G.; Cheze, L.; Armand, S. Feasibility and reliability of using an exoskeleton to emulate muscle contractures during walking. Gait Posture 2016, 50, 239–245. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  4. Wu, Q.; Wang, X.; Du, F.; Xi, R. Modeling and position control of a therapeutic exoskeleton targeting upper extremity rehabilitation. Proc. IMechE Part C J. Mech. Eng. Sci. 2017, 231, 4360–4373. [Google Scholar] [CrossRef]
  5. Zhou, J.; Yang, S.; Xue, Q. Lower limb rehabilitation exoskeleton robot: A review. Adv. Mech. Eng. 2021, 13, 1–17. [Google Scholar] [CrossRef]
  6. Young, A.J.; Ferris, D.P. State of the art and future directions for lower limb robotic exoskeletons. IEEE T. Neur. Sys. Reh. 2017, 25, 171–182. [Google Scholar] [CrossRef]
  7. Babič, J.; Laffranchi, M.; Tessari, F.; Verstraten, T.; Novak, D.; Šarabon, N.; Ugurlu, B.; Peternel, L.; Torricelli, D.; Veneman, J.F. Challenges and solutions for application and wider adoption of wearable robots. Wearable Technol. 2021, 2, e14. [Google Scholar] [CrossRef]
  8. Asl, H.J.; Yamashita, M.; Narikiyo, T.; Kawanishi, M. Field-based assist-as-needed control schemes for rehabilitation robots. IEEE-ASME T. Mech. 2020, 25, 2100–2111. [Google Scholar] [CrossRef]
  9. Wang, D.; Lee, K.M.; Guo, J.; Yang, C.J. Adaptive knee joint exoskeleton based on biological geometries. IEEE-ASME T. Mech. 2014, 19, 1268–1278. [Google Scholar] [CrossRef]
  10. Yan, Y.; Liu, G.; Zhang, L.; Gong, R.; Fu, P.; Han, B.; Li, H. Biomechanical effect of valgus knee braces on the treatment of medial gonarthrosis: A systematic review. Appl. Bionics Biomech. 2022, 4194472, 1–15. [Google Scholar] [CrossRef]
  11. Kim, J.H.; Shim, M.; Ahn, D.H.; Son, B.J.; Kim, S.-Y.; Kim, D.Y.; Baek, Y.S.; Cho, B.-K. Design of a knee exoskeleton using foot pressure and knee torque sensors. Int. J. Adv. Robot Syst. 2015, 12, 112. [Google Scholar] [CrossRef]
  12. Edrisi, K. Designing a backstepping sliding mode controller for an assistant human knee exoskeleton based on nonlinear disturbance observer. Mechatronics 2018, 54, 121–132. [Google Scholar]
  13. Chen, B.; Zi, B.; Wang, Z.; Qin, L.; Liao, W.-H. Knee exoskeletons for gait rehabilitation and human performance augmentation: A state-of-the-art. Mech. Mach. Theory 2019, 134, 499–511. [Google Scholar] [CrossRef]
  14. Mikolajczyk, T.; Ciobanu, I.; Badea, D.I.; Iliescu, A.; Pizzamiglio, S.; Schauer, T.; Seel, T.; Seiciu, P.L.; Turner, D.L.; Berteanu, M. Advanced technology for gait rehabilitation: An overview. Adv. Mech. Eng. 2018, 10, 1–19. [Google Scholar] [CrossRef]
  15. Wang, Y.; Zhang, W.; Shi, D.; Geng, Y. Design and control of an adaptive knee joint exoskeleton mechanism with buffering function. Sensors 2021, 21, 8390. [Google Scholar] [CrossRef]
  16. Karavas, N.; Ajoudani, A.; Tsagarakis, N.; Saglia, J.; Bicchi, A.; Caldwell, D. Tele-impedance based assistive control for a compliant knee exoskeleton. Robot Auton. Syst. 2015, 73, 78–90. [Google Scholar] [CrossRef]
  17. Shi, D.; Zhang, W.; Zhang, W.; Ding, X. A review on lower limb rehabilitation exoskeleton robots. Chin. J. Mech. Eng-En. 2019, 32, 1–11. [Google Scholar] [CrossRef] [Green Version]
  18. de Andrade, R.M.; Ulhoa, P.H.F.; Dias, E.A.F.; Filho, A.B.; Vimieiro, C.B.S. Design and testing a highly backdrivable and kinematic compatible magneto-rheological knee exoskeleton. J. Intel. Mat. Syst. Str. 2022. [Google Scholar] [CrossRef]
  19. Kardan, I.; Akbarzadeh, A. Robust output feedback assistive control of a compliantly actuated knee exoskeleton. Robot Auton. Syst. 2017, 98, 15–29. [Google Scholar] [CrossRef]
  20. Li, F.; Wang, Q.; Xie, Y.; Xie, H. Admittance control of four-link bionic knee exoskeleton with inertia compensation. Tech. Gaz. 2019, 27, 891–897. [Google Scholar]
  21. Li, Z.; Ma, W.; Yin, Z.; Guo, H. Tracking control of time-varying knee exoskeleton disturbed by interaction torque. ISA Trans. 2017, 71, 458–466. [Google Scholar] [CrossRef] [PubMed]
  22. Sanchez-Villamañan, M.D.; Gonzalez-Vargas, J.; Torricelli, D.; Moreno, J.C.; Pons, J.L. Compliant lower limb exoskeletons: A comprehensive review on mechanical design principles. J. Neuroeng. Rehabil. 2019, 16, 55. [Google Scholar] [CrossRef]
  23. Tiboni, M.; Borboni, A.; Vérité, F.; Bregoli, C.; Amici, C. Sensors and actuation technologies in exoskeletons: A review. Sensors 2022, 22, 884. [Google Scholar] [CrossRef] [PubMed]
  24. Tucker, M.R.; Shirota, C.; Lambercy, O.; Sulzer, J.S.; Gassert, R. Design and characterization of an exoskeleton for perturbing the knee during gait. Trans. Biomed. Eng. 2017, 64, 2331–2343. [Google Scholar] [CrossRef]
  25. Lee, K.M.; Guo, J. Kinematic and dynamic analysis of an anatomically based knee joint. J. Biomech. 2010, 43, 1231–1236. [Google Scholar] [CrossRef]
  26. Zoss, A.B.; Kazerooni, H.; Chu, A. Biomechanical design of the Berkeley lower extremity exoskeleton (BLEEX). IEEE/ASME T. Mech. 2006, 11, 128–138. [Google Scholar] [CrossRef]
  27. Zhang, L.; Liu, G.; Han, B.; Wang, Z.; Li, H.; Jiao, Y. Assistive devices of human knee joint: A review. Robot Auton. Syst. 2020, 125, 103394. [Google Scholar] [CrossRef]
  28. Mavroidis, C.; Nikitczuk, J.; Weinberg, B.; Arango, R.; Danaher, G.; Jensen, K.; Leahey, M.; Pavone, R.; Pelletier, P.; Provo, A.; et al. Smart portable rehabilitation devices. In Proceedings of the International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, Long Beach, CA, USA, 24–28 September 2005; pp. 501–510. [Google Scholar]
  29. Pratt, J.E.; Krupp, B.T.; Morse, C.J.; Collins, S.H. The RoboKnee: An exoskeleton for enhancing strength and endurance during walking. In Proceedings of the 2017 IEEE International Conference on Robotics and Automation (ICRA), New Orleans, LA, USA, 26 April–01 May 2004; pp. 2430–2435. [Google Scholar]
  30. Rifaï, H.; Mohammed, S.; Hassani, W.; Amirat, Y. Nested saturation based control of an actuated knee joint orthosis. Mechatronics 2013, 23, 1141–1149. [Google Scholar] [CrossRef]
  31. Shamaei, K.; Napolitano, P.C.; Dollar, A.M. Design and functional evaluation of a quasi-passive compliant stance control knee–ankle–foot orthosis. IEEE T. Neur. Sys. Reh. 2014, 22, 258–268. [Google Scholar] [CrossRef]
  32. Long, Y.; Peng, Y. Design and Control of a Quasi-direct Drive Actuated Knee Exoskeleton. J. Bionic Eng. 2022, 19, 678–687. [Google Scholar] [CrossRef]
  33. Yan, L.; Fan, L.; Xiao, J.; Wang, F. Dynamics analysis and simulation verification of a novel knee joint exoskeleton. J. Vibroeng. 2017, 19, 3008–3018. [Google Scholar]
  34. Elliott, G.; Marecki, A.; Herr, H. Design of a clutch–spring knee exoskeleton for running. J. Med. Devices 2014, 8, 031002. [Google Scholar] [CrossRef] [Green Version]
  35. Fisher, S.; Lucas, L.; Thrasher, T.A. Robot-assisted gait training for patients with hemiparesis due to stroke Top. Stroke Rehabil. 2011, 18, 269–276. [Google Scholar] [CrossRef]
  36. Riener, R.; Lünenburger, L.; Colombo, G. Human-centered robotics applied to gait training and assessment. J. Rehabil. Res. Dev. 2006, 43, 679–694. [Google Scholar] [CrossRef] [PubMed]
  37. Veneman, J.F.; Kruidhof, R.; Hekman, E.E.; Ekkelenkamp, R.; Van Asseldonk, E.H.; Van Der Kooij, H. Design and evaluation of the LOPES exoskeleton robot for interactive gait rehabilitation. IEEE Trans. Neural Syst. Rehabil. Eng. 2007, 15, 379–386. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  38. Ayad, S.; Ayad, M.; Megueni, A.; Spaich, E.G.; Struijk, L.N.S.A. Toward standardizing the classification of robotic gait rehabilitation systems. IEEE Rev. Biomed. Eng. 2018, 12, 138–153. [Google Scholar] [CrossRef] [Green Version]
  39. Saccares, L.; Sarakoglou, I.; Tsagarakis, N.G. iT-Knee: An exoskeleton with ideal torque transmission interface for ergonomic power augmentation. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots & Systems, Daejeon, Republic of Korea, 9–14 October 2016; pp. 780–786. [Google Scholar]
  40. Huber, M.; Eschbach, M.; Kazerounian, K.; Ilies, H.T. Functional evaluation of a personalized orthosis for knee osteoarthritis: A motion capture analysis. J. Med. Devices 2021, 15, 041003. [Google Scholar] [CrossRef]
  41. Choi, B.; Lee, Y.; Kim, Y.J.; Lee, J.; Lee, M.; Roh, S.G.; Park, Y.J.; Kim, K.; Shim, Y. Development of adjustable knee joint for walking assistance devices. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Vancouver, BC, Canada, 24–28 September 2017; pp. 1790–1797. [Google Scholar]
  42. Zhu, Y.; Nakamura, M.; Ito, N.; Fujimoto, H.; Horikuchi, K.; Wakabayashi, S.; Takahashi, R.; Terada, H.; Haro, H. Study of wearable knee assistive instruments for walk rehabilitation. J. Adv. Mech. Des. Syst. 2012, 6, 260–273. [Google Scholar] [CrossRef]
  43. Lee, Y.; Lee, J.; Choi, B.; Lee, M.; Roh, S.-G.; Kim, K.; Seo, K.; Kim, Y.-J.; Shim, Y. Flexible gait enhancing mechatronics system for lower limb assistance (GEMS L.-TYPE). IEEE/ASME T. Mech. 2019, 24, 1520–1531. [Google Scholar] [CrossRef]
  44. Wang, J.; Li, X.; Huang, T.-H.; Yu, S.; Li, Y.; Chen, T.; Carriero, A.; Oh-Park, M.; Su, H. Comfort-centered design of a lightweight and backdrivable knee exoskeleton. IEEE Robotic Autom. Let. 2018, 3, 4265–4272. [Google Scholar] [CrossRef]
  45. Ergin, M.A.; Patoglu, V.A. self-adjusting knee exoskeleton for robot-assisted treatment of knee injuries. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, San Francisco, CA, USA, 25–30 September 2011; pp. 4917–4922. [Google Scholar]
  46. Celebi, B.; Yalcin, M.; Patoglu, V. AssistON-knee: A self-aligning knee exoskeleton. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, Tokyo, Japan, 03–07 November 2013. [Google Scholar]
  47. Kim, T.; Jeong, M.; Kong, K. Bio-inspired knee joint of a lower-limb exoskeleton for misalignment reduction. IEEE/ASME T. Mech. 2022, 27, 1223–1232. [Google Scholar] [CrossRef]
  48. Moon, D.H.; Kim, D.; Hong, Y.D. Development of a single leg knee exoskeleton and sensing knee center of rotation change for intention detection. Sensors 2019, 19, 3960. [Google Scholar] [CrossRef] [Green Version]
  49. Li, H.; Sui, D.; Ju, H.; An, Y.; Zhao, J.; Zhu, Y. Mechanical compliance and dynamic load isolation design of lower limb exoskeleton for locomotion assistance. IEEE/ASME T. Mech. 2022, 27, 5392–5402. [Google Scholar] [CrossRef]
  50. Caulcrick, C.; Huo, W.; Hoult, W.; Vaidyanathan, R. Human joint torque modelling with MMG and EMG during lower limb human-exoskeleton interaction. IEEE Robot Automa. Let. 2021, 6, 7185–7192. [Google Scholar] [CrossRef]
  51. Lee, T.; Lee, D.; Song, B.; Baek, Y.S. Design and control of a polycentric knee exoskeleton using an electro-hydraulic actuator. Sensors 2019, 20, 211. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  52. Meng, Q.; Zeng, Q.; Xie, Q.; Fei, C.; Kong, B.; Lu, X.; Wang, H.; Yu, H. Flexible lower limb exoskeleton systems: A review. NeuroRehab. 2022, 50, 367–390. [Google Scholar] [CrossRef]
  53. Petsch, S.; Rix, R.; Khatri, B.; Schuhladen, S.; Müller, P.; Zentel, R.; Zappe, H. Smart artificial muscle actuators: Liquid crystal elastomers with integrated temperature feedback. Sens. Actuators A: Phys. 2015, 231, 44–51. [Google Scholar] [CrossRef]
  54. Lee, D.; Song, B.; Park, S.Y.; Baek, Y.S. Development and control of an electro-hydraulic actuator system for an exoskeleton robot. Appl. Sci. 2019, 9, 4295. [Google Scholar] [CrossRef] [Green Version]
  55. Zhu, J.; Wang, Y.; Jiang, J.; Sun, B.; Cao, H. Unidirectional variable stiffness hydraulic actuator for load-carrying knee exoskeleton. Int. J. Adv. Robot Syst. 2017, 14, 1–12. [Google Scholar] [CrossRef] [Green Version]
  56. Kaminaga, H.; Amari, T.; Niwa, Y.; Nakamura, Y. Development of knee power assist using backdrivable electro-hydrostatic actuator. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, Taipei, Taiwan, 18–22 October 2010; pp. 5517–5524. [Google Scholar]
  57. Klein Rot, J.E. Evaluation of a force controlled pneumatically actuated knee orthosis. Master’s Thesis, University of Twente, Enschede, The Netherlands, 2021. [Google Scholar]
  58. Aoyagi, D.; Ichinose, W.E.; Harkema, S.J.; Reinkensmeyer, D.J.; Bobrow, J.E. A robot and control algorithm that can synchronously assist in naturalistic motion during body-weight-supported gait training following neurologic injury. IEEE T. Neur. Sys. Reh. 2007, 15, 387–400. [Google Scholar] [CrossRef]
  59. Zhao, W.; Song, A. Active motion control of a knee exoskeleton driven by antagonistic pneumatic muscle actuators. Actuators 2020, 9, 134. [Google Scholar] [CrossRef]
  60. Xiloyannis, M.; Alicea, R.; Georgarakis, A.-M.; Haufe, F.L.; Wolf, P.; Masia, L.; Riener, R. Soft robotic suits: State of the art, core technologies, and open challenges. IEEE T. Robot 2022, 38, 1343–1362. [Google Scholar] [CrossRef]
  61. Veneman, J.F.; Ekkelenkamp, R.; Kruidhof, R.; van der Helm, F.C.; van der Kooij, H. A series elastic-and bowden-cable-based actuation system for use as torque actuator in exoskeleton-type robots. Int. J. Robot. Res. 2006, 25, 261–281. [Google Scholar] [CrossRef]
  62. Jain, P.; Bera, T.K.; Singla, A.; Isaksson, M. Linear actuator–based knee exoskeleton for stand–sit–stand motions: A bond graph approach. Simul-T Soc. Mod. Sim. 2022, 98, 627–644. [Google Scholar] [CrossRef]
  63. Horst, R.W.; Marcus, R.R. Flexcva: A continuously variable actuator for active orthotics. In Proceedings of the International Conference of the IEEE Engineering in Medicine and Biology Society, New York, NY, USA, 30 August–3 September 2006; pp. 2425–2428. [Google Scholar]
  64. Kong, K.; Bae, J.; Tomizuka, M. A compact rotary series elastic actuator for human assistive systems. IEEE/ASME T. Mech. 2011, 17, 288–297. [Google Scholar] [CrossRef]
  65. Chen, G.; Qi, P.; Guo, Z.; Yu, H. Mechanical design and evaluation of a compact portable knee–ankle–foot robot for gait rehabilitation. Mech. Mach. Theory 2016, 103, 51–64. [Google Scholar] [CrossRef]
  66. Zhou, C.; Li, C.; Song, Y.; Lei, Y.; Wang, J.; Wang, C.; Zeng, F. Optimal design and command filtered backstepping control of exoskeleton with series elastic actuator. J. Dyn. Syst. Meas. Control 2022, 144, 091002. [Google Scholar] [CrossRef]
  67. Zhu, L.; Shi, X.; Chen, Z.; Zhang, H.-T.; Xiong, C.-H. Adaptive servomechanism of pneumatic muscle actuators with uncertainties. IEEE Trans. Ind. Electron. 2016, 64, 3329–3337. [Google Scholar] [CrossRef]
  68. Beyl, P.; Knaepen, K.; Duerinck, S.; Van Damme, M.; Vanderborght, B.; Meeusen, R.; Lefeber, D. Safe and compliant guidance by a powered knee exoskeleton for robot-assisted rehabilitation of gait. Adv. Robot. 2011, 25, 513–535. [Google Scholar] [CrossRef]
  69. Maeda, D.; Tominaga, K.; Oku, T.; Pham, H.T.T.; Saeki, S.; Uemura, M.; Hirai, H.; Miyazaki, F. Muscle synergy analysis of human adaptation to a variable-stiffness exoskeleton: Human walk with a knee exoskeleton with pneumatic artificial muscles. In Proceedings of the IEEE-RAS International Conference on Humanoid Robots, Osaka, Japan, 29 November–1 December 2012; pp. 638–644. [Google Scholar]
  70. Mohri, S.; Inose, H.; Yokoyama, K.; Yamada, Y.; Kikutani, I.; Nakamura, T. Development of endoskeleton type knee auxiliary power assist suit using pneumatic artificial muscles. In Proceedings of the IEEE International Conference on Advanced Intelligent Mechatronics (AIM), Banff, AB, Canada, 12–15 July 2016; pp. 107–112. [Google Scholar]
  71. Veale, A.J.; Staman, K.; Van Der Kooij, H. Soft, wearable, and pleated pneumatic interference actuator provides knee extension torque for sit-to-stand. Soft Robot. 2021, 8, 28–43. [Google Scholar] [CrossRef]
  72. Cao, J.; Xie, S.Q.; Das, R. MIMO sliding mode controller for gait exoskeleton driven by pneumatic muscles. IEEE Trans. Control. Syst. Technol. 2017, 26, 274–281. [Google Scholar] [CrossRef] [Green Version]
  73. Aguilar-Sierra, H.; Lopez, R.; Yu, W.; Salazar, S.; Lozano, R. A lower limb exoskeleton with hybrid actuation. In Proceedings of the IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics, Sao Paulo, Brazil, 12–15 August 2014; pp. 695–700. [Google Scholar]
  74. Cao, Y.; Huang, J.; Huang, Z.; Tu, X.; Mohammed, S. Optimizing control of passive gait training exoskeleton driven by pneumatic muscles using switch-mode firefly algorithm. Robotica 2019, 37, 2087–2103. [Google Scholar] [CrossRef]
  75. Sawicki, G.S.; Ferris, D.P. A pneumatically powered knee-ankle-foot orthosis (KAFO) with myoelectric activation and inhibition. J. Neuroeng. Rehabil. 2009, 6, 1–16. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  76. Zhu, Y.; Wu, Q.; Chen, B.; Xu, D.; Shao, Z. Design and evaluation of a novel torque-controllable variable stiffness actuator with reconfigurability. IEEE/ASME T. Mech. 2021, 27, 292–303. [Google Scholar] [CrossRef]
  77. Shamaei, K.; Cenciarini, M.; Adams, A.A.; Gregorczyk, K.N.; Schiffman, J.M.; Dollar, A.M. Biomechanical effects of stiffness in parallel with the knee joint during walking. IEEE Trans. Biomed. Eng. 2015, 62, 2389–2401. [Google Scholar] [CrossRef] [PubMed]
  78. Bacek, T.; Moltedo, M.; Rodriguez-Guerrero, C.; Geeroms, J.; Vanderborght, B.; Lefeber, D. Design and evaluation of a torque-controllable knee joint actuator with adjustable series compliance and parallel elasticity. Mech. Mach. Theory 2018, 130, 71–85. [Google Scholar] [CrossRef]
  79. Tian, M.; Wang, X.; Wang, J.; Gan, Z. Design of a lower limb exoskeleton driven by tendon-sheath artificial muscle. In Proceedings of the IEEE International conference on robotics and biomimetics (ROBIO), Dali, China, 6–8 December 2019; pp. 2037–2042. [Google Scholar]
  80. Chen, B.; Wang, B.; Zheng, C.; Zi, B. Design and simulation of a robotic knee exoskeleton with a variable stiffness actuator for gait rehabilitation. In Proceedings of the IEEE International Conference on Mechatronics and Machine Vision in Practice (M2VIP), Shanghai, China, 26–28 November 2021; pp. 24–29. [Google Scholar]
  81. Cestari, M.; Sanz-Merodio, D.; Garcia, E. A new and versatile adjustable rigidity actuator with add-on locking mechanism (ARES-XL). Actuators 2018, 7, 11–21. [Google Scholar] [CrossRef] [Green Version]
  82. Baček, T.; Moltedo, M.; Geeroms, J.; Vanderborght, B.; Rodriguez-Guerrero, C.; Lefeber, D. Varying mechanical compliance benefits energy efficiency of a knee joint actuator. Mechatronics 2020, 66, 102318. [Google Scholar] [CrossRef]
  83. Zhang, J.; Cong, M.; Liu, D.; Du, Y.; Ma, H. A lightweight variable stiffness knee exoskeleton driven by shape memory alloy. Ind. Robot. Int. J. Robot. Res. Appl. 2022, 49, 994–1007. [Google Scholar] [CrossRef]
  84. Zhang, L.; Huang, Q.; Cai, K.; Wang, Z.; Wang, W.; Liu, J. A wearable soft knee exoskeleton using vacuum-actuated rotary actuator. IEEE Access 2020, 8, 61311–61326. [Google Scholar] [CrossRef]
  85. Li, Y.; Hashimoto, M. Design and prototyping of a novel lightweight walking assist wear using PVC gel soft actuators. Sensors Actuat. A Phys. 2016, 239, 26–44. [Google Scholar] [CrossRef]
  86. Pamungkas, D.S.; Caesarendra, W.; Soebakti, H.; Analia, R.; Susanto, S. Overview: Types of lower limb exoskeletons. Electronics 2019, 8, 1283. [Google Scholar] [CrossRef] [Green Version]
  87. Zhu, Y.; Wu, Q.; Chen, B.; Zhao, Z. Design and voluntary control of variable stiffness exoskeleton based on sEMG driven mode. IEEE Robot Autom. Lett. 2022, 7, 5787–5794. [Google Scholar] [CrossRef]
  88. Gui, K.; Tan, U.-X.; Liu, H.; Zhang, D. Electromyography-driven progressive assist-as-needed control for lower limb exoskeleton. IEEE Trans. Med. Robo. Bionics 2020, 2, 50–58. [Google Scholar] [CrossRef]
  89. Pinto-Fernandez, D.; Torricelli, D.; Sanchez-Villamanan, M.D.C.; Aller, F.; Mombaur, K.; Conti, R.; Vitiello, N.; Moreno, J.C.; Pons, J.L. Performance evaluation of lower limb exoskeletons: A systematic review. IEEE T. Neur. Sys. Reh. 2020, 28, 1573–1583. [Google Scholar] [CrossRef]
  90. Li, W.-Z.; Cao, G.-Z.; Zhu, A.-B. Review on control strategies for lower limb rehabilitation exoskeletons. IEEE Access 2021, 9, 123040–123060. [Google Scholar] [CrossRef]
  91. Chang, Y.J.; Liang, J.N.; Hsu, M.J.; Lien, H.Y.; Fang, C.Y.; Lin, C.H. Effects of continuous passive motion on reversing the adapted spinal circuit in humans with chronic spinal cord injury. Arc. Phys. Med. and Rehab. 2013, 94, 822–828. [Google Scholar] [CrossRef]
  92. Aguirre-Ollinger, G.; Colgate, J.E.; Peshkin, M.A.; Goswami, A. Design of an active one-degree-of-freedom lower-limb exoskeleton with inertia compensation. Ind. Robot 2011, 30, 486–499. [Google Scholar] [CrossRef]
  93. Mohammed, S.; Huo, W.; Huang, J.; Rifaï, H.; Amirat, Y. Nonlinear disturbance observer based sliding mode control of a human-driven knee joint orthosis. Robot Auton. Syst. 2016, 75, 41–49. [Google Scholar] [CrossRef]
  94. Lee, D.; McLain, B.J.; Kang, I.; Young, A.J. Biomechanical comparison of assistance strategies using a bilateral robotic knee exoskeleton. IEEE T. Bio-Med. Eng. 2021, 68, 2870–2879. [Google Scholar] [CrossRef]
  95. Shan, H.; Jiang, C.; Mao, Y.; Wang, X. Design and control of a wearable active knee orthosis for walking assistance. In Proceedings of the IEEE International Workshop on Advanced Motion Control (AMC), Auckland, New Zealand, 22–24 April 2016; pp. 51–56. [Google Scholar]
  96. Shamaei, K.; Cenciarini, M.; Adams, A.A.; Gregorczyk, K.N.; Schiffman, J.M.; Dollar, A.M. Design and evaluation of a quasi-passive knee exoskeleton for investigation of motor adaptation in lower extremity joints. IEEE T. Bio-Med. Eng. 2014, 61, 1809–1821. [Google Scholar] [CrossRef] [PubMed]
  97. Zhou, Z.; Liao, Y.; Wang, C.; Wang, Q. Preliminary evaluation of gait assistance during treadmill walking with a light-weight bionic knee exoskeleton. In Proceedings of the IEEE International Conference on Robotics and Biomimetics (ROBIO), Qingdao, China, 03–07 December 2016; pp. 1173–1178. [Google Scholar]
  98. Pagani, C.H.F.; Potthast, W.; Brüggemann, G.P. The effect of valgus bracing on the knee adduction moment during gait and running in male subjects with varus alignment. Clin. Biomech. 2010, 25, 70–76. [Google Scholar] [CrossRef] [PubMed]
  99. Kamali, K.; Akbari, A.A.; Akbarzadeh, A. Trajectory generation and control of a knee exoskeleton based on dynamic movement primitives for sit-to-stand assistance. Adv. Robot 2016, 30, 846–860. [Google Scholar] [CrossRef]
  100. Pagani, C.H.F.; Willwacher, S.; Kleis, B.; Brüggemann, G.P. Influence of a valgus knee brace on muscle activation and co-contraction in patients with medial knee osteoarthritis. J. Electromyogr. and Kinesiol. 2013, 23, 490–500. [Google Scholar] [CrossRef]
  101. Galle, S.; Malcolm, P.; Derave, W.; De Clercq, D. Adaptation to walking with an exoskeleton that assists ankle extension. Gait Posture 2013, 38, 495–499. [Google Scholar] [CrossRef] [PubMed]
  102. Sun, Y.; Tang, Y.; Zheng, J.; Dong, D.; Chen, X.; Bai, L. From sensing to control of lower limb exoskeleton: A systematic review. Annu. Rev. Control 2022, 53, 83–96. [Google Scholar] [CrossRef]
Applsci 13 01064 g001 550
Figure 1. Human knee joint and its movement: (a) anatomical diagram of the knee joint; (b) knee polycentric motion in the sagittal plane.
Figure 1. Human knee joint and its movement: (a) anatomical diagram of the knee joint; (b) knee polycentric motion in the sagittal plane.
Applsci 13 01064 g001
Applsci 13 01064 g002 550
Figure 2. Knee exoskeletons with a simplified structure design: (a) the ERF based knee device; (b) the RoboKnee exoskeleton; (c) the EICOSI orthosis; (d) the KAFO; (e) the knee exoskeleton using IMU sensors.
Figure 2. Knee exoskeletons with a simplified structure design: (a) the ERF based knee device; (b) the RoboKnee exoskeleton; (c) the EICOSI orthosis; (d) the KAFO; (e) the knee exoskeleton using IMU sensors.
Applsci 13 01064 g002
Applsci 13 01064 g003 550
Figure 3. Knee exoskeletons based on biological geometries: (a) the knee exoskeleton with multi-bar mechanism; (b) the GEMS using cam and other mechanisms; (c) the personalized knee exoskeleton combined with belt and other mechanical parts; (d) the assisted knee exoskeleton with an under-actuated Schmidt coupling; (e) the bioinspired knee joint with a curved guide rail and three ligament bearings.
Figure 3. Knee exoskeletons based on biological geometries: (a) the knee exoskeleton with multi-bar mechanism; (b) the GEMS using cam and other mechanisms; (c) the personalized knee exoskeleton combined with belt and other mechanical parts; (d) the assisted knee exoskeleton with an under-actuated Schmidt coupling; (e) the bioinspired knee joint with a curved guide rail and three ligament bearings.
Applsci 13 01064 g003
Applsci 13 01064 g004 550
Figure 4. Knee exoskeletons with conventional driving actuators: (a) the knee exoskeleton actuated by a linear hydraulic; (b) the active knee exoskeleton utilizing a pneumatic cylinder actuator; (c) the sensor-based knee exoskeleton actuated by a linear actuator; (d) the wearable exoskeleton with a servo motor driven module actuator; (e) the compliant knee exoskeletons with series elastic actuator.
Figure 4. Knee exoskeletons with conventional driving actuators: (a) the knee exoskeleton actuated by a linear hydraulic; (b) the active knee exoskeleton utilizing a pneumatic cylinder actuator; (c) the sensor-based knee exoskeleton actuated by a linear actuator; (d) the wearable exoskeleton with a servo motor driven module actuator; (e) the compliant knee exoskeletons with series elastic actuator.
Applsci 13 01064 g004
Applsci 13 01064 g005 550
Figure 5. Knee exoskeletons with pneumatic muscle actuators: (a) the powered exoskeleton actuated by two antagonistic pleated PAMs; (b) the robotic knee exoskeleton with soft McKibben PAMs; (c) the power assist suit actuated by straight-fiber-type artificial muscles; (d) the flexible wearable exoskeleton with a novel pleated pneumatic interference actuator; (e) the compliant exoskeleton system actuated by artificial pneumatic muscles.
Figure 5. Knee exoskeletons with pneumatic muscle actuators: (a) the powered exoskeleton actuated by two antagonistic pleated PAMs; (b) the robotic knee exoskeleton with soft McKibben PAMs; (c) the power assist suit actuated by straight-fiber-type artificial muscles; (d) the flexible wearable exoskeleton with a novel pleated pneumatic interference actuator; (e) the compliant exoskeleton system actuated by artificial pneumatic muscles.
Applsci 13 01064 g005
Applsci 13 01064 g006 550
Figure 6. Knee exoskeletons with variable stiffness actuators: (a) the wearer’s knee joint equipped with two springs; (b) the knee exoskeleton employing two springs working both in series and in parallel; (c) the wearable exoskeleton actuated by tendon-sheath artificial muscles; (d) the robotic knee exoskeleton with a variable stiffness actuator.
Figure 6. Knee exoskeletons with variable stiffness actuators: (a) the wearer’s knee joint equipped with two springs; (b) the knee exoskeleton employing two springs working both in series and in parallel; (c) the wearable exoskeleton actuated by tendon-sheath artificial muscles; (d) the robotic knee exoskeleton with a variable stiffness actuator.
Applsci 13 01064 g006
Table
Table 1. Comparison of the simplified structure and complex structure of powered knee exoskeletons.
Table 1. Comparison of the simplified structure and complex structure of powered knee exoskeletons.
Simplified StructureComplex Structure
construction of knee components(a) a hinge with single degree of freedom;(a) multi-bar mechanism;
(b) modular actuator with rotary motion;(b) a pulley and several rolling cams;
(c) self-adapting joint with Schmidt coupling, guide rail, bearing, etc.;
advantages(a) Easy to design and light in weight;(a) little or no effect on kinematics;
(b) kinematic analysis is very simple;(b) adapt to the knee joint motion;
(c) high accuracy of mechanical motion;(c) suitable for highly recovery patients;
disadvantages(a) alignment and adjustability are difficult;(a) require complex mechanisms;
(b) usually alter normal walking kinematics;(b) can be bulky and high inertia;
(c) may cause discomfort due to undesired constraints and forces;(c) kinematic analysis is difficult;
Table
Table 2. Comparison results of the actuation designs of existing powered knee exoskeletons.
Table 2. Comparison results of the actuation designs of existing powered knee exoskeletons.
Type of ActuatorDriving DevicesAdvantagesDisadvantages
Conventional driving model(a) linear hydraulic actuators;(a) fast response speed;(a) with large stiffness;
(b) linear pneumatic cylinder;(b) easy to control;(b) poor human-machine interaction;
(c) motor in combination with
(d) other transmission mechanisms;		(c) large drive and auxiliary torque;(c) defects in the weight and drive flexibility;
Pneumatic muscles (a) antagonistic pleated PAMs;(a) simple structure;(a) slow response time;
(b) McKibben style PAMs;(b) compliant driving;(b) fixed movement;
(c) straight-fiber-type PAMs;(c) small motion inertia;(c) low control precision;
Variable stiffness actuators(a) adopted the combination of series and parallel springs;(a) behaves similar to biological muscles;(a) multiple components and complex structure;
(b) clutchable elastic device based on Hill muscle model;(b) adaptable to multiple application;(b) the drive system is relatively bulky;
(c) multi-configuration elastic driving model;(c) friendly to user;(c) lack of robustness;
Other novel actuators(a) shape memory alloy;(a) soft, light and flexible;(a) geometries must be exact;
(b) plasticized polyvinyl chloride;(b) biomimicry;(b) difficult to control;
(c) other novel function material;(c) high force/weight ratio;(c) nonlinear hysteresis effect
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Wu, Z.; Yang, M.; Xia, Y.; Wang, L. Mechanical Structural Design and Actuation Technologies of Powered Knee Exoskeletons: A Review. Appl. Sci. 2023, 13, 1064. https://doi.org/10.3390/app13021064

AMA Style

Wu Z, Yang M, Xia Y, Wang L. Mechanical Structural Design and Actuation Technologies of Powered Knee Exoskeletons: A Review. Applied Sciences. 2023; 13(2):1064. https://doi.org/10.3390/app13021064

Chicago/Turabian Style

Wu, Zongpeng, Mingxing Yang, Yulei Xia, and Liwei Wang. 2023. "Mechanical Structural Design and Actuation Technologies of Powered Knee Exoskeletons: A Review" Applied Sciences 13, no. 2: 1064. https://doi.org/10.3390/app13021064

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop