Call for Speakers 2025

EMPOWER reimagines human-machine synergy by empowering machines with advanced intelligence and humans with augmented physical strength. Through the integration of real-time AI into active exoskeletons, EMPOWER creates a human-machine system where human expertise and decision making directs exoskeleton support to optimally augment human physical strength. Multi-modal, multi-sensor perception enables the prediction of workflows, the understanding of complex tasks, and the proactive adjustment of exoskeleton support characteristics. We will present work-in-progress results on multi-modal activity recognition for exoskeleton control. EMPOWER will demonstrate its innovations in industrial environments such as steel processing, logistics and construction, where prototypes with showcase how adaptive body-worn exoskeletons can reduce physical strain, mitigate the risk of musculoskeletal disorders, and preserve human-centered decision-making in demanding work settings.

At the research group Mobilab&Care, we collaborate with the healthcare industry to implement digital design workflows and 3D-printing techniques for the production of customized assistive devices, including orthotics, prosthetics, exoskeleton interfaces, and occupational therapy aids. We present a parametric design approach using Rhino 3D and Grasshopper algorithms to create devices based on 3D scans and anatomical landmarks, enabling optimal fit, improved alignment, and reduced production time. The method supports design flexibility and integration of mechanical components. To address durability concerns, we developed custom mechanical testing protocols, as existing standards are limited for evaluating patient-specific, additively manufactured devices

Manual labor is still strongly present in industrial contexts. this commonly involves tasks requiring working in non-ergonomic conditions and manipulating heavy parts, which can lead to musculoskeletal disorders. the back is one of the most affected regions. the eit-m superhuman project aims to industrialize, test, validate, and assess a hip-low back exoskeleton for labor applications, assisting the workers. the consortium integrates the supsi back-support exoskeleton and the gogoa hip exoskeleton. ctag provides expertise in laboratory tests, and art-er assists the consortium in engaging industrial partners for testing purposes. bonfiglioli riduttori and mch sonae are the two end users testing the proposed solution. in this presentation, we will provide the audience with the design and evaluation strategies adopted in the project.

The presentation focuses on reducing the mass of the rehabilitation exoskeleton of an upper extremity targeted at convenient home use with the possibility of assembling the device to the objects of daily use (such as chairs or beds). the presented investigation included a hybrid approach to finite element model topology and parametric optimisation for additively manufactured objects. it elaborates on the different methods of constructing driving systems and passive joints to enable high-stress resistance and stiffness while not increasing mass and providing full mobility of shoulder and elbow joints - critical for the task-oriented treatment. the design process is based on the real-life registered activities of daily life and multibody dynamics computations of the simulated exoskeleton with different extremity models. the works were held within a research project funded by the national center for research and development (lider xiv, contract number lider14/0196/2023).

We will discusses a novel approach in 3D movement analysis in humans (and non-rigid exoskeletons/robotics). Custom ‘UMIMU sensors’ were developed, each comprising a fully-integrated pair of Magnetic Inertial Measurement Unit (MIMU) and UIWB sensors, with an timing-optimizied embedded protocol measuring all distances within an UMIMU swarm. EKF-based position-estimating methods were developed/validated for achieving the accuracy and reproducibility required for improved 3D movement analysis. A study into ranging errors in typical gait analysis conditions were studied. Their mitigating is subject of current research. Also a novel segment calibration method is discussed that merges the worlds of marker-based and MIMU-based movement analysis, facilitating e.g. MIMU-based gait analysis in patients that cannot perform the standard pose/squat tasks required in current MIMU-based gait analysis.

Nonverbal communication plays a crucial role in social interactions, yet visually impaired individuals face significant challenges in perceiving these cues. This talk explores the development of a haptic glove designed to convey emotional and contextual information through vibrotactile feedback. Based on qualitative research and user studies, we examine which information should be transmitted and how emotions can be effectively perceived haptically. The talk will discuss findings from user interviews, prototype development, and initial evaluations, highlighting the potential of haptic technology to enhance accessibility and social inclusion. By bridging the gap between technology and human interaction, this research aims to create more intuitive and meaningful communication experiences for visually impaired users.

Wearable neurorobots are systems designed to interpret human neural patterns and translate them into commands for wearable robotic devices, such as exoskeletons. However, the widespread adoption of neurorobotics for everyday is still limited. This project aims at redefining the neurorobotic system as a multifaceted symbiotic entity consisting of three intelligent agents---user, decoder, and robotic device. By integrating neuroscientific and robotic methodologies, the goal is to create and actively encourage shared interactions among these intelligent agents: on the one hand, the user interacts with the exoskeleton through a hybrid neural interface; on the other hand, the exoskeleton is treated as a semi-autonomous agent capable of contextualizing the user’s commands and adapt the gait trajectory to the environment.

The assistance from powered lower-limb exoskeletons can extend the duration of the rehabilitation sessions for spinal cord injury and stroke patients. The help of an exoskeleton can reduce or delay the development of muscular fatigue as a risk factor for spasticity. The need to avoid the development of fatigue is crucial in rehabilitation because it negatively impacts the patient's progress. In this presentation, we will provide an overview of our approach to assessing muscular fatigue via electromyographic signal spectral analysis. This non-invasive approach aims to detect fatigue in its early stage of development and represents the first step toward a real-time fatigue detection approach that can be applied in clinical settings for the rehabilitation of neurologically impaired patients.

This study aims to develop a hybrid FES-exoskeleton system to enhance neurorehabilitation for lower limb mobility. By incorporating a biomechanical model of the lower limbs, including joint stiffness, the research examines motion dynamics and muscle activation patterns to enable seamless integration of functional electrical stimulation (FES) and exoskeleton support. The system features adaptable control designs tailored to individual users, ensuring a more personalized approach to rehabilitation. Additionally, real-time electromyography (EMG) signals are utilized as feedback to improve the interaction between the user and the system, fostering more efficient and natural movement patterns.

Industrial exoskeletons are gaining acceptance in workplaces to reduce work-related musculoskeletal disorders (WMSDs). Because workers perform diverse tasks, exoskeletons must be versatile. Active exoskeletons offer greater adaptability than passive exoskeletons, allowing for adjusting force assistance based on user activity. A key challenge is to determine the amount of control users should have while ensuring safety and functionality. Advances in virtual augmented and mixed reality can enhance user interaction. This talk presents a VR-based user interface for occupational exoskeletons. The system allows users to manage key functions such as calibration, adjustment, activation, and force modulation. It can automatically adapt force assistance when users handle loads, thereby improving both efficiency and usability.

Problem: “The human body changes shape throughout the day, and prosthetic fittings don’t.” Prosthetic limbs are attached to the residual leg with U-shaped solid structures called sockets. When the leg changes shape, sockets don’t fit anymore and break the skin, making amputees unable to walk for days. Solution: We developed the world’s first robotic liner; Roliner. Roliner is a sleeve-like device worn on the stump before putting the prosthetic limb on. It uses Artificial Intelligence to understand the hourly/daily changes in the stump and adapts to them. Reducing clinical dependency, Roliner’s AI reads real-time sensors’ data between the leg and socket, learns amputees’ comfort preferences via an app; seamlessly and continuously adjust the fitting by inflating/deflating Roliner’s micro-channels. (www.unhindr.com)

The Kickstart® Walk Assist system is a neurorehabilitation device that consists of a hip belt, a multi-joint bracing system and a spring-powered Exotendon. Kickstart provides stability and assistance to lift, swing and guide the leg properly for each step without using electronics. Consecutive clinical studies demonstrated that the Kickstart exoskeletal device had a positive effect in improving walking ability in stroke survivors. In addition, the exoskeletal device could be easily integrated into current therapeutic programs to help patients enhance walking recovery both for hospital- and community-based rehabilitation training.

Recent technological improvements, particularly in mechatronic systems and materials, have advanced the development of lower limb exoskeletons. However, efficient use requires autonomous adaptation, predicting the user's intended motion and biomechanical changes in lower limb joints, i.e., position and torque estimation. For that purpose, recent studies have focused on using electromyography (EMG) signals. By utilizing time and frequency domain features extracted from EMG signals as input, deep and machine learning-based architectures, e.g., artificial neural networks, support vector machines, have showed to be efficient for accurate predictions. Nevertheless, proposed structures can be improved, serving commercialized product adaptation. Within this context, feature extraction and EMG implementation methods require further research studies in order to achieve reduced computational complexity and real-time applicability.

The objective of the work was to integrate a hybrid wearable that could facilitate physical tasks, and also, that could analyze and alert on risks associated to the development of the tasks. After an ergonomics characterization of workstation, a benchmarking of exoskeletons and sensors, and a review of ergonomic analysis methods: EXOSOFT and Xsens dot sensors were selected and integrated. The methodology of risk assessment selected was the OWAS methodology. A mobile application was created, allowing to have a real time feedback of postures and their risk associated. Also, a validation of the system was carried out: 4 users were measured EMG and MoCap at the laboratory, and 10 users measured in real conditions. A subjective questionnaire was also conducted.

There is growing interest to use soft robotic hands for prosthetic purpose due to their low-lost, light-weight, and safe interaction compared to current prosthetic hands based on rigid robotics. Our soft robotic prosthetic hands are 3D-printable with monolithic structures and integrated with soft sensors as part of their mechanical structure. The integration of soft sensors enables direct transition between different hand gestures, which greatly differs from the commercial hands requiring a neutral position (generally hand-opening) before starting a new gesture. Compared to sEMG, pressure-based FMG sensors (patent pending) are advantageous in decoding user intention from muscle activities due to low cost, stable signals, less signal processing, and light-weight. Together with a pFMG-based pattern recognition system, our hands will provide more intuitive and robust control for prosthetic hand users.

Industrial exoskeletons present a potential solution to reduce musculoskeletal disorders (MSD) and the risk of injury at work. Researchers and developers face design challenges to optimize the exoskeleton performance when the user interacts with it. In particular, active industrial exoskeletons introduce a challenge for developers since users must have control of certain domains in the exoskeleton. This academic talk presents the research, design and evaluation of a Human Machine Interface (HMI) called User Command Interface (UCI), this is an enhancement solution to adapt exoskeleton systems variability between tasks and users. The system can solve current interaction problems among developers and users such as secure identification, parameter configuration, signal visualization and storage for user specifications.

A wearable soft-robotic glove, Carbonhand, was tested for its potential effect on hand function when using it as assistive device. We coordinated a multi-center study to provide input to the manufacturer for CE certification under MDR. People with reduced grip strength used the glove at home for 6 weeks, with assessments of hand function, strength, user experience and quality of life before and after glove use. Interim results (n=46) showed that grip strength and hand function improved after glove use, and maintained up and until 4 weeks post-intervention, with positive user experience. It also gave rise to useful input for system improvements. This is promising for extending rehabilitation into people's homes, while offering assistance during daily activities.

This presentation delves into the application of engineering involving the integration of robotics, automation, and AI to automate life sciences experiments. We propose the utilization of AI methods within a robot-assisted platform for generating three-dimensional tumor tissue models. Recent technological advancements and innovative engineering solutions have the potential of standardization of the automatic production of the models. Offering flexibility in adjusting various experimental parameters. Manually analyzing such data is time-consuming and subject to inaccuracies stemming from individual subjectivity. Hence, there is a compelling need for a robust and efficient solution to evaluate the morphology of these models. Particularly for quality assessment and assessing therapeutic effects for individualized patient therapy.

Children with Cerebral Palsy (CP) mainly suffer from reduced quality of life resulting from diminishing mobility and independence. Recent studies have shown that Lower-limb Exoskeletons (LLEs) have significant potential to improve the walking functionality of children with CP. We developed a 6-DOF LLE, KAROL, with actuated joints in the knee, hip frontal, and hip sagittal. Human gait in LLE has highly unstable dynamics similar to the extremely unstable inverted pendulum. Hence, a control structure is highly desirable to stabilise human-LLE interaction. Furthermore, individuals with CP have more metabolic consumption than unimpaired peers, with gravity significantly contributing to this issue. Therefore, this study designed a model-based gravity compensator impedance controller for KAROL to mitigate the negative impact of gravity on this assistive technology.

SafeLegs is a lower-limb exoskeleton prototype equipped with Deep Learning-based safety functions. First, we employed Reinforcement Learning for critical fault search in a Digital Twin of the exoskeleton. Secondly, by using Machine Learning optimisation, we drastically reduced amount of possible data access points from the system. From the optimised set of APs we trained the LSTM-based signal predictor. This model was quantised and deployed on the edge computer that controls the exoskeleton. As the result, the system is capable to successfully detect hardware faults before they can lead to a catastrophic failure while keeping the computational overhead as minimal as possible.

We propose a discussion on the extent to which currently available overground exoskeletons are versatile enough to accommodate specific demands across different clinical conditions leading to gait rehabilitation. We further address the lack of well-established rehabilitation protocols (e.g., dose and dosage), even when considering a single clinical population, and the limited availability of exoskeletons regulated for clinical use in Canada.

Summary A civil engineering company co-designed a robotic exoskeleton with to decrease the employees exposition to high biomechanical constraints related to mechanical or automated tasks. The French Institute of Health and Safety (INRS) is following this company by for 4 years (2018-2022) aiming to understand and optimize the process of integration and employees’ acceptance of this specific device. This follow-up was carried out using the INRS tools, which are available to companies and occupational health specialists. This longitudinal follow-up highlighted the importance of scientific and technical education about this type of technology at different levels of the organization: individual, collective and organizational. Our communication will consist of presenting the results of the exoskeleton integration project follow-up.

In the application-oriented R&D project ExoPflege, an assistive occupational exoskeleton with active shoulder-arm and adaptive trunk module is under development to support nursing staff members in hospitals. The primary goal is to support manual pushing and pulling motions during bed transfer of narcotized patients, an activity that was identified as physically demanding in a preliminary on-site analysis. A biomechanical model-based approach was used to analyze use-cases, as well as to select and optimize the exoskeleton design. The presentation will outline the application requirements, design process and the current status of the prototype development.

Gait Analysis (GAn) is crucial for precision rehabilitation and clinical decisions. Traditionally conducted via tests and visual assessments, GAn measures spatio-temporal and kinematic parameters to detect gait abnormalities. Optical motion capture systems offer accuracy but are costly and lab-restricted. Marker-less motion analysis (MMA) represents a flexible, marker-free alternative, enabling faster, easier assessments and outdoor use, beneficial for sports biomechanics and daily monitoring of neuromotor disorders. MEMENTO develops a cost-effective, marker-less GAn software, harnessing pose estimation for motion detection using a simple camera, providing real-time feedback on walking performance. This innovation aims to improve the feasibility and usability of GAn in clinical settings, offering a unique training program for healthcare professionals globally in advanced gait assessment.