Hierarchical learning control with physical human-exoskeleton interaction

Review of Vision-Based Environmental Perception for Lower-Limb Exoskeleton Robots

The exoskeleton robot is a wearable electromechanical device inspired by animal exoskeletons. It combines technologies such as sensing, control, information, and mobile computing, enhancing human physical abilities and assisting in rehabilitation training. In recent years, with the development of visual sensors and deep learning, the environmental perception of exoskeletons has drawn widespread attention in the industry. Environmental perception can provide exoskeletons with a certain level of autonomous perception and decision-making ability, enhance their stability and safety in complex environments, and improve the human-machine-environment interaction loop. This paper provides a review of environmental perception and its related technologies of lower-limb exoskeleton robots. First, we briefly introduce the visual sensors and control system. Second, we analyze and summarize the key technologies of environmental perception, including related datasets, detection of critical terrains, and environment-oriented adaptive gait planning. Finally, we analyze the current factors limiting the development of exoskeleton environmental perception and propose future directions.

Advancements in Sensor Technologies and Control Strategies for Lower-Limb Rehabilitation Exoskeletons: A Comprehensive Review

Lower-limb rehabilitation exoskeletons offer a transformative approach to enhancing recovery in patients with movement disorders affecting the lower extremities. This comprehensive systematic review delves into the literature on sensor technologies and the control strategies integrated into these exoskeletons, evaluating their capacity to address user needs and scrutinizing their structural designs regarding sensor distribution as well as control algorithms. The review examines various sensing modalities, including electromyography (EMG), force, displacement, and other innovative sensor types, employed in these devices to facilitate accurate and responsive motion control. Furthermore, the review explores the strengths and limitations of a diverse array of lower-limb rehabilitation-exoskeleton designs, highlighting areas of improvement and potential avenues for further development. In addition, the review investigates the latest control algorithms and analysis methods that have been utilized in conjunction with these sensor systems to optimize exoskeleton performance and ensure safe and effective user interactions. By building a deeper understanding of the diverse sensor technologies and monitoring systems, this review aims to contribute to the ongoing advancement of lower-limb rehabilitation exoskeletons, ultimately improving the quality of life for patients with mobility impairments.

Shaping high-performance wearable robots for human motor and sensory reconstruction and enhancement

Most wearable robots such as exoskeletons and prostheses can operate with dexterity, while wearers do not perceive them as part of their bodies. In this perspective, we contend that integrating environmental, physiological, and physical information through multi-modal fusion, incorporating human-in-the-loop control, utilizing neuromuscular interface, employing flexible electronics, and acquiring and processing human-robot information with biomechatronic chips, should all be leveraged towards building the next generation of wearable robots. These technologies could improve the embodiment of wearable robots. With optimizations in mechanical structure and clinical training, the next generation of wearable robots should better facilitate human motor and sensory reconstruction and enhancement.

Human-in-the-Loop Optimization of Wearable Robotic Devices to Improve Human-Robot Interaction: A Systematic Review

This article presents a systematic review on wearable robotic devices that use human-in-the-loop optimization (HILO) strategies to improve human-robot interaction. A total of 46 HILO studies were identified and divided into upper and lower limb robotic devices. The main aspects from HILO were identified, reviewed, and classified in four areas: 1) human-machine systems; 2) optimization methods; 3) control strategies; and 4) experimental protocols. A variety of objective functions (physiological, biomechanical, and subjective), optimization strategies, and optimized control parameters configurations used in different control strategies are presented and analyzed. An overview of experimental protocols is provided, including metrics, tasks, and conditions tested. Moreover, the relevance given to training or adaptation periods was explored. We outline an HILO framework that includes current wearable robots, optimization strategies, objective functions, control strategies, and experimental protocols. We conclude by highlighting current research gaps and defining future directions to improve the development of advanced HILO strategies in upper and lower limb wearable robots.

Interaction learning control with movement primitives for lower limb exoskeleton

Research on robotic exoskeletons both in the military and medical fields has rapidly expanded over the previous decade. As a human-robot interaction system, it is a challenge to develop an assistive strategy that makes the exoskeleton supply efficient and natural assistance following the user's intention. This paper proposed a novel interaction learning control strategy for the lower extremity exoskeleton. A powerful representative tool probabilistic movement primitives (ProMPs) is adopted to model the motion and generate the desired trajectory in real-time. To adjust the trajectory by the user's real-time intention, a compensation term based on human-robot interaction force is designed and merged into the ProMPs model. Then, compliant impedance control is adopted as a low-level control where the desired trajectory is put into. Moreover, the model will be dynamically adapted online by penalizing both the interaction force and trajectory mismatch, with all the parameters that can be further learned by learning algorithm PI^BB. The experimental results verified the effectiveness of the proposed control framework.

An Improvement of Robot Stiffness-Adaptive Skill Primitive Generalization Using the Surface Electromyography in Human-Robot Collaboration

Learning from Demonstration in robotics has proved its efficiency in robot skill learning. The generalization goals of most skill expression models in real scenarios are specified by humans or associated with other perceptual data. Our proposed framework using the Probabilistic Movement Primitives (ProMPs) modeling to resolve the shortcomings of the previous research works; the coupling between stiffness and motion is inherently established in a single model. Such a framework can request a small amount of incomplete observation data to infer the entire skill primitive. It can be used as an intuitive generalization command sending tool to achieve collaboration between humans and robots with human-like stiffness modulation strategies on either side. Experiments (human–robot hand-over, object matching, pick-and-place) were conducted to prove the effectiveness of the work. Myo armband and Leap motion camera are used as surface electromyography (sEMG) signal and motion capture sensors respective in the experiments. Also, the experiments show that the proposed framework strengthened the ability to distinguish actions with similar movements under observation noise by introducing the sEMG signal into the ProMP model. The usage of the mixture model brings possibilities in achieving automation of multiple collaborative tasks.

Exoskeleton Active Walking Assistance Control Framework Based on Frequency Adaptive Dynamics Movement Primitives

This paper introduces a novel exoskeleton active walking assistance control framework based on frequency adaptive dynamics movement primitives (FADMPs). The FADMPs proposed in this paper is an online learning and prediction algorithm which is able to online estimate the fundamental frequency of human joint trajectory, learn the shape of joint trajectory and predict the future joint trajectory during walking. The proposed active walking assistance control framework based on FADMPs is a model-based controller which relies on the human joint torque estimation. The assistance torque provided by exoskeleton is estimated by human lower limb inverse dynamics model which is sensitive to the noise in the joint motion trajectory. To estimate a smooth joint torque profile, the joint motion trajectory must be filtered first by a lowpass filter. However, lowpass filter will introduce an inevitable phase delay in the filtered trajectory. Both simulations and experiments in this paper show that the phase delay has a significant effect on the performance of exoskeleton active assistance. The active assistant control framework based on FADMPs aims at improving the performance of active assistance control by compensating the phase delay. Both simulations and experiments on active walking assistance control show that the performance of active assistance control can be further improved when the phase delay in the filtered trajectory is compensated by FADMPs.

A Semi-active Exoskeleton Based on EMGs Reduces Muscle Fatigue When Squatting

In dynamic manufacturing and warehousing environments, the work scene made it impossible for workers to sit, so workers suffer from muscle fatigue of the lower limb caused by standing or squatting for a long period of time. In this paper, a semi-active exoskeleton used to reduce the muscle fatigue of the lower limb was designed and evaluated. (i) Background: The advantages and disadvantages of assistive exoskeletons developed for industrial purposes were introduced. (ii) Simulation: The process of squatting was simulated in the AnyBody.7.1 software, the result showed that muscle activity of the gluteus maximus, rectus femoris, vastus medialis, vastus lateralis, vastus intermedius, and erector spinae increased with increasing of knee flexion angle. (iii) Design: The exoskeleton was designed with three working modes: rigid-support mode, elastic-support mode and follow mode. Rigid-support mode was suitable for scenes where the squatting posture is stable, while elastic-support mode was beneficial for working environments where the height of squatting varied frequently.The working environments were identified intelligently based on the EMGs of the gluteus maximus, and quadriceps, and the motor was controlled to switch the working mode between rigid-support mode and elastic-support mode. In follow mode, the exoskeleton moves freely with users without interfering with activities such as walking, ascending and descending stairs. (iv) Experiments: Three sets of experiments were conducted to evaluate the effect of exoskeleton. Experiment one was conducted to measure the surface electromyography signal (EMGs) in both condition of with and without exoskeleton, the root mean square of EMGs amplitude of soleus, vastus lateralis, vastus medialis, gastrocnemius, vastus intermedius, rectus femoris, gluteus maximus, and erector spinae were reduced by 98.5, 97.89, 80.09, 77.27, 96.73, 94.17, 70.71, and 36.32%, respectively, with the assistance of the exoskeleton. The purpose of experiment two was aimed to measure the plantar pressure with and without exoskeleton. With exoskeleton, the percentage of weight through subject's feet was reduced by 63.94, 64.52, and 65.61% respectively at 60°, 90°, and 120° of knee flexion angle, compared to the condition of without exoskeleton. Experiment three was purposed to measure the metabolic cost at a speed of 4 and 5 km/h with and without exoskeleton. Experiment results showed that the average additional metabolic cost introduced by exoskeleton was 2.525 and 2.85%. It indicated that the exoskeleton would not interfere with the movement of the wearer Seriously in follow mode. (v) Conclusion: The exoskeleton not only effectively reduced muscle fatigue, but also avoided interfering with the free movement of the wearer.

A safe human–robot interactive control structure with human arm movement detection for an upper-limb wearable robot used during lifting tasks

Manual lifting tasks involve repetitive raising, holding and stacking movements with heavy objects. These arm movements are notable risk factors for muscle pain, fatigue, and musculoskeletal disorders in workers. An upper-limb wearable robot, as a 6-DOF dual-arm exoskeleton, which was designed to augment workers’ strength and minimize muscular activation in the arm during repetitive lifting tasks. To adjust the robot joint trajectory, the user needs to apply an interactive torque to operate the robot during lifting tasks when a standard virtual mechanical impedance control structure is used. To reduce overshooting of the interactive torque on the user’s joint, a three-tier hierarchical control structure was developed for the robot in this study. At the highest level, a human arm movement detection module is used to detect the user’s arm motion according to the surface electromyography signals. Then, a Hammerstein adaptive virtual mechanical impedance controller is used at the middle level to reduce overshooting and yield an acceptable value of torque for the user’s elbow joint in actual lifting tasks. At the lowest level, the actuator controller on each joint of the robot controls the robot to complete lifting tasks. Several experiments were conducted, and the results showed that the interactive torque on the user’s elbow was limited and the muscular activations of erector spinae and biceps brachii muscles were effectively decreased. The proposed scheme prevents potential harm to the user due to excessive interactive torque on the human elbow joint, such as related muscle fatigue and joint injuries.

An adaptive stair-ascending gait generation approach based on depth camera for lower limb exoskeleton

The mobility on stairways is a daily challenge for seniors and people with dyskinesia. Lower limb exoskeletons can be effective assistants to improve their life quality. In this paper, we present an adaptive stair-ascending gait generation algorithm based on a depth camera for lower limb exoskeletons. We first construct a linked-list-based stairway model with the point cloud captured from the depth camera. Then, an optimal foothold point is calculated based on the linked-list stair model for gait generation. Finally, the exoskeleton takes the stair-ascending gait of healthy people as a reference and generates appropriate gait for the stair. The proposed gait generation algorithm is initially validated through holistic simulation analyses. We tested the stairway modeling algorithm on varieties of indoor and outdoor stairways and evaluated the gait generation algorithm on stairs of different height. The subjects' stair walking tests with lower limb exoskeletons show the effectiveness of the proposed stairway modeling and gait generation approaches.

Dissociating motor learning from recovery in exoskeleton training post-stroke

Background

A large number of robotic or gravity-supporting devices have been developed for rehabilitation of upper extremity post-stroke. Because these devices continuously monitor performance data during training, they could potentially help to develop predictive models of the effects of motor training on recovery. However, during training with such devices, patients must become adept at using the new “tool” of the exoskeleton, including learning the new forces and visuomotor transformations associated with the device. We thus hypothesized that the changes in performance during extensive training with a passive, gravity-supporting, exoskeleton device (the Armeo Spring) will follow an initial fast phase, due to learning to use the device, and a slower phase that corresponds to reduction in overall arm impairment. Of interest was whether these fast and slow processes were related.

Methods

To test the two-process hypothesis, we used mixed-effect exponential models to identify putative fast and slow changes in smoothness of arm movements during 80 arm reaching tests performed during 20 days of exoskeleton training in 53 individuals with post-acute stroke.

Results

In line with our hypothesis, we found that double exponential models better fit the changes in smoothness of arm movements than single exponential models. In contrast, single exponential models better fit the data for a group of young healthy control subjects. In addition, in the stroke group, we showed that smoothness correlated with a measure of impairment (the upper extremity Fugl Meyer score - UEFM) at the end, but not at the beginning, of training. Furthermore, the improvement in movement smoothness due to the slow component, but not to the fast component, strongly correlated with the improvement in the UEFM between the beginning and end of training. There was no correlation between the change of peaks due to the fast process and the changes due to the slow process. Finally, the improvement in smoothness due to the slow, but not the fast, component correlated with the number of days since stroke at the onset of training – i.e. participants who started exoskeleton training sooner after stroke improved their smoothness more.

Conclusions

Our results therefore demonstrate that at least two processes are involved in in performance improvements measured during mechanized training post-stroke. The fast process is consistent with learning to use the exoskeleton, while the slow process independently reflects the reduction in upper extremity impairment.

Exoskeleton robot control for synchronous walking assistance in repetitive manual handling works based on dual unscented Kalman filter

Prolong walking is a notable risk factor for work-related lower-limb disorders (WRLLD) in industries such as agriculture, construction, service profession, healthcare and retail works. It is one of the common causes of lower limb fatigue or muscular exhaustion leading to poor balance and fall. Exoskeleton technology is seen as a modern strategy to assist worker’s in these professions to minimize or eliminate the risk of WRLLDs. Exoskeleton has potentials to benefit workers in prolong walking (amongst others) by augmenting their strength, increasing their endurance, and minimizing high muscular activation, resulting in overall work efficiency and productivity. Controlling exoskeleton to achieve this purpose for able-bodied personnel without impeding their natural movement is, however, challenging. In this study, we propose a control strategy that integrates a Dual Unscented Kalman Filter (DUKF) for trajectory generation/prediction of the spatio-temporal features of human walking (i.e. joint position, and velocity, and acceleration) and an impedance cum supervisory controller to enable the exoskeleton to follow this trajectory to synchronize with the human walking. Experiment is conducted with four subjects carrying a load and walking at their normal speed- a typical scenario in industries. EMG signals taken at two muscles: Right Vastus Intermedius (on the thigh) and Right Gastrocnemius (on the calf) indicated reduction in muscular activation during the experiment. The results also show the ability of the control system to predict spatio-temporal features of the pilots’ walking and to enable the exoskeleton to move in concert with the pilot.