ABSTRACT
Introduction: Lower limb exoskeletons have shown considerable potential in assisting human walking, particularly by reducing metabolic cost (MC), leading to a surge of interest in this field in recent years. However, owing to significant individual differences and the uncertainty of movements, challenges still exist in the personalized design and control of exoskeletons in human-robot interactions. Methods: In this study, we propose a hybrid data-driven approach that integrates musculoskeletal simulation with machine learning technology to customize personalized assistance strategies efficiently and adaptively for ankle-foot exoskeletons. First, optimal assistance strategies that can theoretically minimize MC, were derived from forward muscle-driven simulations on an open-source dataset. Then, a neural network was utilized to explore the relationships among different individuals, movements, and optimal strategies, thus developing a predictive model. Results: With respect to transfer learning, our approach exhibited effectiveness and adaptability when faced with new individuals and movements. The simulation results further indicated that our approach successfully reduced the MC of calf muscles by approximately 20% compared to normal walking conditions. Discussion: This hybrid approach offers an alternative for personalizing assistance strategy that may further guide exoskeleton design.
ABSTRACT
A biologically-inspired actuation system, including muscles, spinal reflexes, and vestibular feedback, may be capable of achieving more natural gait mechanics in powered prostheses or exoskeletons. In this study, we developed a Virtual Muscle Reflex (VMR) system to control ankle torque and tuned it using data from human responses to anteroposterior mechanical perturbations at three walking speeds. The system consists of three Hill-Type muscles, simulated in real time, and uses feedback from ground reaction force and from stretch sensors on the virtual muscle fibers. Controller gains, muscle properties, and reflex/vestibular time delays were optimized using Covariance Matrix Adaptation (CMA) to minimize the difference between the VMR torque output and the torque measured from the experiment. We repeated the procedure using a conventional finite-state impedance controller. For both controllers, the coefficient of determination (R2) and root-mean-square error (RMSE) was calculated as a function of time within the gait cycle. The VMR had lower RMSE than the impedance controller in 70%, and in 60% of the trials, the R2 of the VMR controller was higher than for the impedance controller. We concluded that the VMR system can better reproduce the human responses to perturbations than the impedance controller.
ABSTRACT
Exoskeletons are technologies that can help to increase or improve mobility, dexterity, and strength. They can be used as assistive devices, to restore lost affordances, or for rehabilitation. While mechanical exoskeletons are passive and rely on the body's power for movement, powered exoskeletons are active mechanical systems that can assist or enhance a user's capacity, including in strength and performance. They also offer scope to augment or enhance beyond simple medical support, with potential in the future for superhuman power and strength. While these technologies present promising clinical opportunities, including for those who want to regain walking capacity, they also bring ethical questions, such as about data privacy and accessibility. In addition, the physical features of the technology can prove mentally, physically, and financially demanding, and may be deployed in contexts where user choice and autonomy is constrained. In this article, we discuss these issues, and raise some pertinent ethical questions, not all of which can be easily answered. We touch upon medical and therapeutic uses, for industrial and workplace settings, and in military contexts specially, given these are contexts where such technology may be required or even imposed. We argue that reasonable optimism for such technologies needs to be tempered by sufficient ethical assessment to identify and address barriers to research, development, and use. As well as managing any impacts and expectations for the health and wellbeing of users, the potential impact on autonomy and the risk of coercion, we have to consider what kind of data may be recorded or used, and the risk that these technologies could exacerbate existing inequalities or harms.
ABSTRACT
The research community has conducted several controlled "in -lab" assessments on the effectiveness of industrial exoskeletons, paving the way for their adoption. However, field testing, focusing on ergonomics and the user experience, could serve to enhance both end-users' awareness and address open doubts concerning true effectiveness of industrial exoskeletons. This study presents an analysis of qualitative data regarding the use of back-support exoskeletons during field trials in harsh civil engineering environments. This work evaluates the StreamEXO's (an active back-support exoskeleton) efficacy in reducing fatigue and the evolution of its perceived usefulness. This is achieved using qualitative data collection tools, during real scenarios testing over multiple-day trials. Collected data shows a positive correlation between self-reported fatigue, measured on a four verbal anchors-based Borg CR10 scale, and the use of the exoskeleton during physically demanding movements. Moreover, the evolution of scores throughout the testing sessions (90 minutes of exoskeleton use for three nonconsecutive days) suggests a trend due to the adaptation and learning curve of workers during the exoskeleton experience. The analysis of the open-ended answers highlights that the adaptation to physical interaction has a negative oscillation on day two to rise back during the third day, possibly correlated to a change in muscle pattern. The main critical factors affecting comfort during the exoskeleton experience are weight balance, body pressure, and thermal comfort, which can strongly affect device acceptance.
ABSTRACT
In order to better design handling-assisted exoskeletons, it is necessary to analyze the biomechanics of human hand movements. In this study, Anybody Modeling System (AMS) simulation was used to analyze the movement state of muscles during human handling. Combined with surface electromyography (sEMG) experiments, specific analysis and verification were carried out to obtain the position of muscles that the human body needs to assist during handling. In this study, the simulation and experiment were carried out for the manual handling process. A treatment group and an experimental group were set up. This study found that the vastus medialis muscle, vastus lateralis muscle, latissimus dorsi muscle, trapezius muscle, deltoid muscle and triceps brachii muscle require more energy in the process of handling, and it is reasonable and effective to combine sEMG signals with the simulation of the musculoskeletal model to analyze the muscle condition of human movement.
Subject(s)
Electromyography , Exoskeleton Device , Muscle, Skeletal , Humans , Electromyography/methods , Muscle, Skeletal/physiology , Biomechanical Phenomena/physiology , Movement/physiology , Male , Adult , Hand/physiologyABSTRACT
As both the proportion of older people and the length of life increases globally, a rise in age-related degenerative diseases, disability, and prolonged dependency is projected. However, more sophisticated biomedical materials, as well as an improved understanding of human disease, is forecast to revolutionize the diagnosis and treatment of conditions ranging from osteoarthritis to Alzheimer's disease as well as impact disease prevention. Another, albeit quieter, revolution is also taking place within society: human augmentation. In this context, humans seek to improve themselves, metamorphosing through self-discipline or more recently, through use of emerging medical technologies, with the goal of transcending aging and mortality. In this review, and in the pursuit of improved medical care following aging, disease, disability, or injury, we first highlight cutting-edge and emerging materials-based neuroprosthetic technologies designed to restore limb or organ function. We highlight the potential for these technologies to be utilized to augment human performance beyond the range of natural performance. We discuss and explore the growing social movement of human augmentation and the idea that it is possible and desirable to use emerging technologies to push the boundaries of what it means to be a healthy human into the realm of superhuman performance and intelligence. This potential future capability is contrasted with limitations in the right-to-repair legislation, which may create challenges for patients. Now is the time for continued discussion of the ethical strategies for research, implementation, and long-term device sustainability or repair.
ABSTRACT
Wearable ExoNETs offer a novel, wearable solution to support and facilitate upper extremity gravity compensation in healthy, unimpaired individuals. In this study, we investigated the safety and feasibility of gravity compensating ExoNETs on 10 healthy, unimpaired individuals across a series of tasks, including activities of daily living and resistance exercises. The direct muscle activity and kinematic effects of gravity compensation were compared to a sham control and no device control. Mixed effects analysis revealed significant reductions in muscle activity at the biceps, triceps and medial deltoids with effect sizes of -3.6%, -4.5%, and -7.2% rmsMVC, respectively, during gravity support. There were no significant changes in movement kinematics as evidenced by minimal change in coverage metrics at the wrist. These findings reveal the potential for the ExoNET to serve as an alternative to existing bulky and encumbering devices in post-stroke rehabilitation settings and pave the way for future clinical trials.
ABSTRACT
BACKGROUND: Wearable trunk exoskeletons hold immense potential in fields such as healthcare and industry. Previous research has indicated that intention recognition control plays a crucial role in users' daily use of exoskeletons. OBJECTIVE: This review aims to discuss the characteristics of intention recognition control schemes for intelligent trunk exoskeletons under different control objectives over the past decade. METHODS: Considering the relatively late development of active trunk exoskeletons, we selected papers published in the last decade (2013 to 2023) from the Web of Science, PubMed, and IEEE Xplore databases. In total, 50 articles were selected and examined based on four control objectives. RESULTS: In general, we found that researchers focus on trunk exoskeleton devices designed for assistance and motor augmentation, which rely more on body movement signals as a source for intention recognition. CONCLUSION: Based on these results, we identify and discuss several promising research directions that may help to attain a widely accepted control methods, thereby advancing further development of trunk exoskeleton technology.
Subject(s)
Exoskeleton Device , Intention , Torso , Humans , Torso/physiology , Movement/physiology , Wearable Electronic DevicesABSTRACT
OCCUPATIONAL APPLICATIONSOccupational exoskeletons receive rising interest in industry as these devices diminish the biomechanical load during manual materials handling. Still, we have limited knowledge when it comes to in-field use. This gap often contributes to failure in the implementation of exoskeleton in industry. In this study, we investigated how a training protocol consisting of in-field use of a passive back exoskeleton affected the biomechanics of logistic workers. More specifically, we focused on how the variation of the muscular and kinematic patterns of the user was altered after exoskeleton training. We found that training had a positive effect on exoskeleton use, as a relative decrease of 6-9% in peak back muscle activity was observed post-training. Additionally, training decreased knee flexion by 6°-16°, indicating a more stoop lifting technique. The findings point at the potential benefits of applying a training approach when implementing a back-supporting exoskeleton in logistics.
Background: Occupational exoskeletons are an attractive solution to reduce the prevalence of attrition and work-related musculoskeletal disorders, such as low back pain, among manual workers. However, research has mostly focused on acute effects, while the effects of in-field use, and exoskeleton training are still to be addressed. Purpose: The aim of the present paper was to investigate how in-field use and exoskeleton training affected the biomechanics, acceptance, and comfort of logistic workers when using a passive back exoskeleton. Methods: Twenty workers were randomly distributed into control and intervention group. The tests consisted of standard lifting tasks with and without exoskeleton before and after a 5-week period. The intervention group underwent a 5-week progressive training protocol aiming at increasing the duration of use of the exoskeleton. The variation in muscle activity (surface electromyography) and full-body kinematics (IMU-based motion capture) were assessed during logistic work tasks. Additionally, acceptance, comfort, and perceived effort were collected. Compliance to the training protocol reached 74%. Results: Using the exoskeleton resulted in a 1320% reduced variation in muscle activity of the back muscles across groups and lifting conditions including trunk extension. The changes in variation were driven by a decrease in peak muscle activity, which was further lowered by 69% after the 5-week training. Additionally, training induced decreased knee flexion indicating a more stoop lifting technique in the intervention group. Conclusions: The present results demonstrate that exoskeleton training optimized the human-exoskeleton interaction by deriving more effects of the exoskeleton in this case by lowering the peak muscle activity of the user during manual materials handling. This underlines the importance of introducing training when implementing exoskeletons in industry. Additionally, the results indicate that a progressive implementation of back supporting exoskeletons in logistics can be beneficial in terms of lowering the biomechanical load during manual materials handling.
ABSTRACT
Lower limb exoskeletons have the potential to mitigate work-related musculoskeletal disorders; however, they often lack user-oriented control strategies. Human-in-the-loop (HITL) controls adapt an exoskeleton's assistance in real time, to optimize the user-exoskeleton interaction. This study presents a HITL control for a knee exoskeleton using a CMA-ES algorithm to minimize the users' physical effort, a parameter innovatively evaluated using the interaction torque with the exoskeleton (a muscular effort indicator) and metabolic cost. This work innovates by estimating the user's metabolic cost within the HITL control through a machine-learning model. The regression model estimated the metabolic cost, in real time, with a root mean squared error of 0.66 W/kg and mean absolute percentage error of 26% (n = 5), making faster (10 s) and less noisy estimations than a respirometer (K5, Cosmed). The HITL reduced the user's metabolic cost by 7.3% and 5.9% compared to the zero-torque and no-device conditions, respectively, and reduced the interaction torque by 32.3% compared to a zero-torque control (n = 1). The developed HITL control surpassed a non-exoskeleton and zero-torque condition regarding the user's physical effort, even for a task such as slow walking. Furthermore, the user-specific control had a lower metabolic cost than the non-user-specific assistance. This proof-of-concept demonstrated the potential of HITL controls in assisted walking.
Subject(s)
Algorithms , Exoskeleton Device , Torque , Humans , Knee/physiology , Machine Learning , Male , Muscle, Skeletal/physiology , Adult , Biomechanical Phenomena/physiology , Energy Metabolism/physiology , Walking/physiology , Knee Joint/physiologyABSTRACT
Pediatric gait rehabilitation and guidance strategies using robotic exoskeletons require a controller that encourages user volitional control and participation while guiding the wearer towards a stable gait cycle. Virtual constraint-based controllers have created stable gait cycles in bipedal robotic systems and have seen recent use in assistive exoskeletons. This paper evaluates a virtual constraint-based controller for pediatric gait guidance through comparison with a traditional time-dependent position tracking controller on a newly developed exoskeleton system. Walking experiments were performed with a healthy child subject wearing the exoskeleton under proportional-derivative control, virtual constraint-based control, and while unpowered. The participant questionnaires assessed the perceived exertion and controller usability measures, while sensors provided kinematic, control torque, and muscle activation data. The virtual constraint-based controller resulted in a gait similar to the proportional-derivative controlled gait but reduced the variability in the gait kinematics by 36.72% and 16.28% relative to unassisted gait in the hips and knees, respectively. The virtual constraint-based controller also used 35.89% and 4.44% less rms torque per gait cycle in the hips and knees, respectively. The user feedback indicated that the virtual constraint-based controller was intuitive and easy to utilize relative to the proportional-derivative controller. These results indicate that virtual constraint-based control has favorable characteristics for robot-assisted gait guidance.
ABSTRACT
Chitin, the most prevalent polymer in nature, a significant structural polysaccharide that comes in second only to cellulose. Chitin is a crucial component of fungal cell walls and also present in many other creatures, such as viruses, plants, animals, insect exoskeletons, and crustacean shells. Chitin presents itself as a promising target for the development of biopesticides. It focuses on unraveling the unique structures and biochemical pathways associated with chitin, aiming to identify vulnerabilities that can be strategically leveraged for effective and environmentally sustainable pest control. It involves a comprehensive analysis of chitinase enzymes, chitin biosynthesis, and chitin-related processes across diverse organisms. By elucidating the molecular intricacies involved in chitin metabolism, this review seeks to unveil potential points of intervention that can disrupt essential biological processes in target pests without harming non-target species. This holistic approach to understanding chitin-related pathways aims to inform the design and optimization of biopesticides with enhanced specificity and reduced ecological impact. The outcomes of this study hold great promise for advancing innovative and eco-friendly pest management strategies. By targeting chitin structures and pathways, biopesticides developed based on these findings may offer a sustainable and selective alternative to conventional chemical pesticides, contributing to the ongoing efforts towards more environmentally conscious and effective pest control solutions.
Subject(s)
Chitin , Chitinases , Chitin/metabolism , Chitin/chemistry , Animals , Chitinases/metabolism , Chitinases/chemistry , Biological Control Agents/metabolism , Biological Control Agents/chemistry , Pest Control, Biological/methods , Insecta/metabolism , Fungi/metabolism , Pesticides/chemistry , Pesticides/metabolismABSTRACT
Accurate prediction of human locomotion intent benefits the seamless switching of lower limb exoskeleton controllers in different terrains to assist humans in walking safely. In this paper, a deep belief network (DBN) was developed to construct a multimodal framework for recognizing various locomotion modes and predicting transition tasks. Three fusion strategies (data level, feature level, and decision level) were explored, and optimal network performance was obtained. This method could be tested on public datasets. For the continuous performance of steady state, the best prediction accuracy achieved was 97.64% in user-dependent testing and 96.80% in user-independent testing. During the transition state, the system accurately predicted all transitions (user-dependent: 96.37%, user-independent: 95.01%). The multimodal framework based on DBN can accurately predict the human locomotion intent. The experimental results demonstrate the potential of the proposed model in the volition control of the lower limb exoskeleton.
ABSTRACT
Commonly used risk indexes, such as the NIOSH Lifting Index, do not capture the effect of exoskeletons. This makes it difficult for Health and Safety professionals to rigorously assess the benefit of such devices. The community requires a simple method to assess the effectiveness of back-support exoskeleton's (BSE) in possibly reducing ergonomic risk. The method introduced in this work is termed "Equivalent Weight" (EqW) and it proposes an interpretation of the effect built on the benefit delivered through reduced activation of the erector spinae (ES). This manifests itself as an apparent reduction of the lifted load perceived by the wearer. This work presents a pilot study where a practical application of the EqW method is used to assess the ergonomic risk in manual material handling (MMH) when using a back support exoskeleton (StreamEXO). The results are assessed by combining observational measurements from on-site testing with five different workers and quantitative measures of the muscle activity reduction achieved during laboratory evaluation with ten workers. These results will show that when lifting, lowering, and carrying a 19 kg load the StreamEXO can reduce risk by up to two levels (from "high" to "low") in the target sub-tasks. The Lifting index (LI) was reduced up to 64% when examining specific sub-tasks and the worker's movement conduction.
Subject(s)
Electromyography , Ergonomics , Exoskeleton Device , Lifting , Railroads , Task Performance and Analysis , Weight-Bearing , Humans , Male , Pilot Projects , Adult , Weight-Bearing/physiology , Ergonomics/methods , Back Muscles/physiology , Female , Risk Assessment/methods , Middle AgedABSTRACT
The global aging population faces significant health challenges, including an increasing vulnerability to disability due to natural aging processes. Wearable lower limb exoskeletons (LLEs) have emerged as a promising solution to enhance physical function in older individuals. This systematic review synthesizes the use of LLEs in alignment with the WHO's healthy aging vision, examining their impact on intrinsic capacities and functional abilities. We conducted a comprehensive literature search in six databases, yielding 36 relevant articles covering older adults (65+) with various health conditions, including sarcopenia, stroke, Parkinson's Disease, osteoarthritis, and more. The interventions, spanning one to forty sessions, utilized a range of LLE technologies such as Ekso®, HAL®, Stride Management Assist®, Honda Walking Assist®, Lokomat®, Walkbot®, Healbot®, Keeogo Rehab®, EX1®, overground wearable exoskeletons, Eksoband®, powered ankle-foot orthoses, HAL® lumbar type, Human Body Posturizer®, Gait Enhancing and Motivation System®, soft robotic suits, and active pelvis orthoses. The findings revealed substantial positive outcomes across diverse health conditions. LLE training led to improvements in key performance indicators, such as the 10 Meter Walk Test, Five Times Sit-to-Stand test, Timed Up and Go test, and more. Additionally, enhancements were observed in gait quality, joint mobility, muscle strength, and balance. These improvements were accompanied by reductions in sedentary behavior, pain perception, muscle exertion, and metabolic cost while walking. While longer intervention durations can aid in the rehabilitation of intrinsic capacities, even the instantaneous augmentation of functional abilities can be observed in a single session. In summary, this review demonstrates consistent and significant enhancements in critical parameters across a broad spectrum of health conditions following LLE interventions in older adults. These findings underscore the potential of LLE in promoting healthy aging and enhancing the well-being of older adults.
Subject(s)
Exoskeleton Device , Healthy Aging , Humans , Aged , Postural Balance , Time and Motion Studies , World Health OrganizationABSTRACT
Most lower limb rehabilitation robots are limited to specific training postures to adapt to stroke patients in multiple stages of recovery. In addition, there is a lack of attention to the switching functions of the training side, including left, right, and bilateral, which enables patients with hemiplegia to rehabilitate with a single device. This article presents an exoskeleton robot named the multistage hemiplegic lower-limb rehabilitation robot, which has been designed to do rehabilitation in multiple training postures and training sides. The mechanism consisting of the thigh, calf, and foot is introduced. Additionally, the design of the multi-mode limit of the hip, knee, and ankle joints supports delivering therapy in any posture and training sides to aid patients with hemiplegia in all stages of recovery. The gait trajectory is planned by extracting the gait motion trajectory model collected by the motion capture device. In addition, a control system for the training module based on adaptive iterative learning has been simulated, and its high-precision tracking performance has been verified. The gait trajectory experiment is carried out, and the results verify that the trajectory tracking performance of the robot has good performance.
Subject(s)
Hemiplegia , Robotics , Humans , Lower Extremity , Foot , GaitABSTRACT
Spinal cord injury (SCI) negatively impacts individuals' functional independence, and motor and sensory function. Intense walking training has been shown to facilitate recovery for individuals with chronic SCI. Powered robotic exoskeletons provide therapists with a tool that allows them to conduct walking training with less therapist effort as compared to conventional walking training. Exoskeletal-assisted walking (EAW) has been studied in the chronic SCI population with preliminary reports showing benefits in mobility, health, and quality-of-life outcomes. However, few reports have studied EAW's benefits in the acute (<90 days post) SCI population at a time when neural plasticity is most dynamic and modifiable. The purpose of the study was to conduct a pilot randomized controlled trial (RCT) to understand the effects of incorporated EAW in acute inpatient rehabilitation (AIR) for individuals with SCI on functional, motor, and sensory recovery. The study outcomes included the Spinal Cord Independence Measure (SCIM) III and International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) motor and sensory scores that were assessed by unblinded assessors. We also recorded EAW session data, including adverse events, walking and up time, step counts, Borg Rating of Perceived Exertion (RPE), and compliance with scheduled EAW training. From August 2019 to July 2022, 16 participants completed the AIR with incorporated EAW, and 12 completed the standard AIR, all with SCI and preserved leg function within 90 days post-injury. During each session, the AIR with incorporated EAW group averaged 34.3 (±9.4) min of up time, 25.4 (±7.7) min of walk time, and 536 (±157) steps. Analysis via two-by-two mixed-effects models showed significant increases in the SCIM total score and ISNCSCI total motor and sensory scores over time for the AIR with incorporated EAW group [SCIM total score: F(1, 26) = 5.59, p = 0.03; total motor score: F(1, 26) = 8.06, p < 0.01; total sensory score: F(1, 19.2) = 5.08, p = 0.04], outperforming the standard AIR group. The AIR with incorporated EAW group showed 13, 14, and 22 points higher changes in the SCIM total score, total motor score, and total sensory score (respectively) by discharge compared with the standard AIR group. Incorporating EAW into AIR may facilitate functional, motor, and sensory recovery for individuals with SCI during AIR better than standard AIR. However, the study had a limited sample size. Further studies are needed to clarify the effects of EAW in AIR.
ABSTRACT
Industrial back support exoskeletons (BSEs) are a promising approach to addressing low back pain (LBP) which still affect a significant proportion of the workforce. They aim to reduce lumbar loading, the main biomechanical risk factor for LBP, by providing external support to the lumbar spine. The aim of this study was to determine the supporting effect of one active (A1) and two passive (P1 and P2) BSEs during different manual material handling tasks. Kinematic data and back muscle activity were collected from 12 subjects during dynamic lifting and static holding of 10 kg. Mean and peak L5/S1 extension moments, L5/S1 compression forces and muscle activation were included in the analysis. During dynamic lifting all BSEs reduced peak (12-26 %) and mean (4-17 %) extension moments and peak (10-22 %) and mean (4-15 %) compression forces in the lumbar spine. The peak (13-28 %) and mean (4-32 %) activity of the back extensor muscles was reduced accordingly. In the static holding task, analogous mean reductions for P1 and P2 of L5/S1 extension moments (12-20 %), compression forces (13-23 %) and muscular activity (16-23 %) were found. A1 showed a greater reduction during static holding for extension moments (46 %), compression forces (41 %) and muscular activity (54 %). This pronounced difference in the performance of the BSEs between tasks was attributed to the actuators used by the different BSEs.
Subject(s)
Lifting , Low Back Pain , Lumbar Vertebrae , Weight-Bearing , Humans , Biomechanical Phenomena , Male , Adult , Weight-Bearing/physiology , Low Back Pain/physiopathology , Lumbar Vertebrae/physiology , Exoskeleton Device , Female , Back Muscles/physiology , Muscle, Skeletal/physiologyABSTRACT
Center of mass (COM) state, specifically in a local reference frame (i.e., relative to center of pressure), is an important variable for controlling and quantifying bipedal locomotion. However, this metric is not easily attainable in real time during human locomotion experiments. This information could be valuable when controlling wearable robotic exoskeletons, specifically for stability augmentation where knowledge of COM state could enable step placement planners similar to bipedal robots. Here, we explored the ability of simulated wearable sensor-driven models to rapidly estimate COM state during steady state and perturbed walking, spanning delayed estimates (i.e., estimating past state) to anticipated estimates (i.e., estimating future state). We used various simulated inertial measurement unit (IMU) sensor configurations typically found on lower limb exoskeletons and a temporal convolutional network (TCN) model throughout this analysis. We found comparable COM estimation capabilities across hip, knee, and ankle exoskeleton sensor configurations, where device type did not significantly influence error. We also found that anticipating COM state during perturbations induced a significant increase in error proportional to anticipation time. Delaying COM state estimates significantly increased accuracy for velocity estimates but not position estimates. All tested conditions resulted in models with R2 > 0.85, with a majority resulting in R2 > 0.95, emphasizing the viability of this approach. Broadly, this preliminary work using simulated IMUs supports the efficacy of wearable sensor-driven deep learning approaches to provide real-time COM state estimates for lower limb exoskeleton control or other wearable sensor-based applications, such as mobile data collection or use in real-time biofeedback.