Michael Sommerhalder

Last Name

Sommerhalder

First Name

Michael

ORCID

0000-0002-3877-3098

Show all metadata

Search Results

Publications 1 - 10 of 11

Therapists’ Force-Profile Teach-and-Mimic Approach for Upper-Limb Rehabilitation Exoskeletons
Luciani, Beatrice; Sommerhalder, Michael; Gandolla, Marta; et al. (2024)
IEEE Transactions on Medical Robotics and Bionics
In this work, we propose a framework enabling upper-limb rehabilitation exoskeletons to mimic the personalised haptic guidance of therapists. Current exoskeletons face acceptability issues as they limit physical interaction between clinicians and patients and offer only predefined levels of support that cannot be tuned during the movements, when needed. To increase acceptance, we first developed a method to estimate the therapist’s force contribution while manipulating a patient’s arm using an upper-limb exoskeleton. We achieved a precision of 0.31Nm without using direct sensors. Then, we exploited the Learning-by-demonstration paradigm to learn from the therapist’s interactions. Single-joint experiments on ANYexo demonstrate that our framework, applying the Vector-search approach, can record the joint-level therapist’s interaction forces during simple tasks, link them to the kinematics of the robot, and then provide support to the user’s limb. The support is coherent with what is learnt and changes with the real-time arm kinematic configuration of the robot, assisting whatever movement the patient executes in the end-effector space without the need for manual regulation. In this way, robotic therapy sessions can exploit therapists’ expertise while reducing their manual workload.
Polymorphic Control Framework for Automated and Individualized Robot-Assisted Rehabilitation
Sommerhalder, Michael; Zimmermann, Yves; Song, Jaeyong; et al. (2024)
IEEE Transactions on Robotics
Robots were introduced in the field of upper-limb neuro-rehabilitation to relieve the therapist from physical labor, and to provide high-intensity therapy to the patient. A variety of control methods were developed that incorporate patients' physiological and biomechanical states to adapt the provided assistance automatically. Higher level states such as selected type of assistance, chosen task characteristics, defined session goals, and given patient impairments are often neglected or modeled into tight requirements, low-dimensional study designs, and narrow inclusion criteria so that presented solutions cannot be transferred to other tasks, robotic devices or target groups. In this work, we present the design of a modular high-level control framework based on invariant states covering all decision layers in therapy. We verified the functionality of our framework on the assistance and task layer by outlaying the invariant states based on the characteristics of twenty examined state-of-the-art controllers. Then, we integrated four controllers on each layer and designed two algorithms that automatically selected suitable controllers. The framework was deployed on an arm rehabilitation robot and tested on one participant acting as a patient. We observed plausible system reactions to external changes by a second operator representing a therapist. We believe that this work will boost the development of novel controllers and selection algorithms in cooperative decision-making on layers other than assistance, and eases transferability and integration of existing solutions on lower layers into arbitrary robotic systems.
Trajectory Optimization Framework for Rehabilitation Robots with Multi-Workspace Objectives and Constraints
Sommerhalder, Michael; Zimmermann, Yves; Simovic, Leonardo; et al. (2023)
IEEE Robotics and Automation Letters
Robot-assisted neurorehabilitation requires trajectories between arbitrary poses in the patient's range of motion. Data-driven optimization methods, such as Learning by Demonstration, are well suited to replicate complex multi-joint movements. However, these methods lack individualization to patient-, robot- and exercise-specific constraints. We propose a hybrid optimization framework that combines cost-based objectives, such as minimizing jerk, with the data-driven optimization of a reference trajectory. The objectives can be individually weighted in a sequential quadratic program with application-related constraints represented in intuitive workspaces. We demonstrated that trajectories recorded from an existing upper-limb activity dataset could be adapted to the personal needs of a healthy participant with simulated impairments, the hardware-specific robot topology, and changes in the exercise setup. Furthermore, we showed how redundancies in the degrees of freedom of the arm can be exploited: For example, an elbow angle movement of 30.4${^\circ }$ was compensated entirely through increased wrist movement in a reach-goal task. In addition to making sequential quadratic programming more accessible to the field of rehabilitation robotics, our framework improves the variability and individualizability of generated trajectories for patients, provides more adaptation possibilities to the therapist, and enables sharing of recorded movement data between robotic platforms, patients, and exercises.
ANYexo 2.0: A Fully-Actuated Upper-Limb Exoskeleton for Manipulation and Joint-Oriented Training in all Stages of Rehabilitation
Zimmermann, Yves; Sommerhalder, Michael; Wolf, Peter; et al. (2023)
IEEE Transactions on Robotics
We developed an exoskeleton for neurorehabilitation that covered all relevant degrees of freedom of the human arm while providing enough range of motion, speed, strength, and haptic-rendering function for therapy of severely affected (e.g., mobilization) and mildly affected patients (e.g., strength and speed). The ANYexo 2.0, uniting these capabilities, could be the vanguard for highly versatile therapeutic robotics applicable to a broad target group and an extensive range of exercises. Thus, supporting the practical adoption of these devices in clinics. The unique kinematic structure of the robot and the bio-inspired controlled shoulder coupling allowed training for most activities of daily living. We demonstrated this capability with 15 sample activities, including interaction with real objects and the own body with the robot in transparent mode. The robot’s joints can reach 200%, 398%, and 354% of the speed required during activities of daily living at the shoulder, elbow, and wrist, respectively. Further, the robot can provide isometric strength training. We present a detailed analysis of the kinematic properties and propose algorithms for intuitive control implementation.
Cooperative Goal Generation for Reaching Tasks in Robot-Assisted Rehabilitation
Sommerhalder, Michael; Zimmermann, Yves; Riener, Robert; et al. (2023)
2023 International Conference on Rehabilitation Robotics (ICORR)
Robot-assisted neurorehabilitation requires automated generation of goal positions for reaching tasks in functional movement therapy. In state-of-the-art solutions, these positions are determined by a motivational therapy game either through constraints on the end-effector (2D or 3D games), or individual arm joints (1D games). Consequently, these positions cannot be adapted to the patients' specific needs by the therapist, and the effectiveness of the training is reduced. We solve this issue by generating goal positions using Gaussian Mixture Models and probability density maps based on the active range of motion of the patient and desired activities, while being compliant with existing game constraints. Therapists can modify the goal generation via an intuitive difficulty and activity parameter. The pipeline was tested on the upper-limb exoskeleton ANYexo 2.0. We have shown that the range of motion exploration rate could be altered from 0.39% to 5.9% per task and that our method successfully generated a sequence of reaching tasks that matched the range of motion of the selected activity, up to an inlier accuracy of 78.9%. Results demonstrate that the responsibilities of the therapy game (i.e., motivating the patient) and the therapists (i.e., individualizing the training) could be distributed properly. We believe that with our pipeline, effective cooperation between the involved agents is achieved, and the provided therapy can be improved.
Score rectification for online assessments in robot-assisted arm rehabilitation
Sommerhalder, Michael; Zimmermann, Yves; Knecht, Manuel; et al. (2022)
at - Automatisierungstechnik ~ Special issue: AUTOMED 2021: Automation in Medical Technology
Relative comparison of clinical scores to measure the effectiveness of neuro-rehabilitation therapy is possible through a series of discrete measurements during the rehabilitation period within specifically designed task environments. Robots allow quantitative, continuous measurement of data. Resulting robotic scores are also only comparable within similar context, e.g. type of task. We propose a method to decouple these scores from their respective context through functional orthogonalization and compensation of the compounding factors based on a data-driven sensitivity analysis of the user performance. The method was validated for the established accuracy score with variable arm weight support, provoked muscle fatigue and different task directions on 6 participants of our arm exoskeleton group on the ANYexo robot. In the best case, the standard deviation of the assessed score in changing context could be reduced by a factor of 3.2. Therewith, we paved the way to context-independent, quantitative online assessments, recorded autonomously with robots.
A Polymorphic Control Framework for Robot-Assisted Neurorehabilitation
Sommerhalder, Michael (2024)
The leading cause of long-term disability and acute hospitalization in high-income countries is stroke, with more than 13.7 million new cases worldwide each year. Stroke often results in paralysis of the limbs. To recover from lost functionality, long-term, intensive neurorehabilitation addressing individualized tasks and active participation of the patient is needed. To relieve the therapists from parts of the physical workload, to increase training efficiency beyond the therapist’s capabilities, and to allow quantitative, continuous measurements of the executed intervention, robots were introduced in the field almost two decades ago. In an ideal setting, therapists and robots collaborate to provide quality-assured intensive therapy for multiple patients. Recent robotic hardware such as the ANYexo 2.0 has the potential to achieve such a collaboration through its versatility, speed, and strength. To exploit the potential of the hardware, a multitude of assistive controllers have been developed. Recent controllers attempt to achieve partially unsupervised therapy by adapting the provided assistance automatically. However, the strong focus on automation of the assistance layer has resulted in a variety of narrow-purpose solutions addressing very specific tasks. Higher layer states such as the selected type of assistance, the chosen task characteristics, the defined session goals, and the given patient impairments have often been neglected or modeled into tight requirements, low-dimensional study designs, and narrow inclusion criteria. Consequently, the presented controllers are hardly transferable to other tasks, robotic devices, or patient and therapist target groups. Due to the versatility of activities performed with the arms and hands, and due to the versatility of comorbidities among patients, narrow-purpose solutions and associated studies lose their significance in the clinical setting. Thus, the overarching objective of this thesis was to develop a unified control framework that is capable of integrating existing and novel specialized control methods, incorporating patients' physiological and biomechanical states, and that allows for automatic or on-demand adjustment of assistance through therapeutic input. This work makes three contributions: Firstly, a polymorphic control framework based on invariant states was developed, which encompasses all decision layers in therapy. The functionality of the framework was verified on the assistance and task layers. Secondly, with the newly established control framework as a basis, we designed layer-specific controllers that can adapt to the broad context of higher layers such as therapy goals, the patient's range of motion, and activities of daily living. In this way, we developed a method for rectifying robotic recordings according to the context in which the data was acquired. This enables therapists to make scores comparable and interpretable between sessions, exercises and tasks. Furthermore, a method was constructed to enable safe haptic interaction with the patient's torso and head, which can adapt to the patient's dynamic body movements. Moreover, we developed methods for generating target positions and trajectories based on a selected therapy goal, and adhere to constraints from the exercise, the patient, and the robot. Thirdly, an observational study was conducted to analyze emerging therapeutic interaction strategies with the objective of tailoring parameters and analysis tools towards the individual preferences of the therapist in order to exploit the increased interaction possibilities that resulted from the layer-specific controllers. Furthermore, we investigated alternative interface options beyond the screen and mouse. We developed an intuitive user interface, ARMStick, which was designed to represent a miniature human arm that can be manipulated by therapists intuitively. Our polymorphic control framework represents a significant advance in the field of software and control for current upper-limb exoskeletons. Unlike previous systems, it addresses the challenges in a holistic manner, taking into account all therapeutic decision layers. This foundation work is intended to facilitate the development of novel controllers and selection algorithms for cooperative decision-making on layers other than assistance. Ultimately, it should facilitate the transferability and integration of existing solutions on lower layers into arbitrary robotic systems.
How the CYBATHLON Competition Has Advanced Assistive Technologies
Jaeger, Lukas; Baptista, Roberto de Souza; Basla, Chiara; et al. (2023)
Annual Review of Control, Robotics, and Autonomous Systems
Approximately 1.1. billion people worldwide live with some form of disability, and assistive technology has the potential to increase their overall quality of life. However, the end users' perspective and needs are often not sufficiently considered during the development of this technology, leading to frustration and nonuse of existing devices. Since its first competition in 2016, CYBATHLON has aimed to drive innovation in the field of assistive technology by motivating teams to involve end users more actively in the development process and to tailor novel devices to their actual daily-life needs. Competition tasks therefore represent unsolved daily-life challenges for people with disabilities and serve the purpose of benchmarking the latest developments from research laboratories and companies from around the world. This review describes each of the competition disciplines, their contributions to assistive technology, and remaining challenges in the user-centered development of this technology.
Digital Guinea Pig: Merits and Methods of Human-in-the-Loop Simulation for Upper-Limb Exoskeletons
Zimmermann, Yves; Sommerhalder, Michael; Song, Jaeyong; et al. (2022)
2022 International Conference on Rehabilitation Robotics (ICORR)
Exoskeletons operate in continuous haptic interaction with a human limb. Thus, this interaction is a key factor to consider during the development of hardware and control policies for these devices. Physics simulations can complement real-world experiments for prototype validation, leading to higher efficiency in hardware and software development iterations as well as increased safety for participants and robot hardware. Here, we present a simulation framework of the full rigid-body dynamics of a coupled human and exoskeleton arm built to validate the full software stack. We present a method to model the human-robot interaction dynamics as decoupled spring-damper systems based on anthropometric data. Further, we demonstrate the application of the simulation framework to predict the closed-loop haptic-rendering performance of a 9-DOF exoskeleton in interaction with a human. The simulation was capable of simulating the closed-loop system's reaction to an impact on a haptic wall. The intrusion into the compliant walls was predicted with a relative accuracy of 6 % to 13 %. Admissible control gains could be predicted with an accuracy of around 14 %, and higher prediction accuracy is indicated when modeling the torque tracking bandwidth of the actuators. Hence, the simulation is valuable for validating prototype software, developing intuition, and a better understanding of the complex characteristics of the coupled system dynamics, even though the quantitative prediction is limited.
Physical Human-Robot Interaction with Real Active Surfaces using Haptic Rendering on Point Clouds
Sommerhalder, Michael; Zimmermann, Yves; Cizmeci, Burak; et al. (2020)
2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
During robot-assisted therapy of hemiplegic patients, interaction with the patient must be intrinsically safe. Straight-forward collision avoidance solutions can provide this safety requirement with conservative margins. These margins heavily reduce the robot’s workspace and make interaction with the patient’s unguided body parts impossible. However, interaction with the own body is highly beneficial from a therapeutic point of view. We tackle this problem by combining haptic rendering techniques with classical computer vision methods. Our proposed solution consists of a pipeline that builds collision objects from point clouds in real-time and a controller that renders haptic interaction. The raw sensor data is processed to overcome noise and occlusion problems. Our proposed approach is validated on the 6 DoF exoskeleton ANYexo for direct impacts, sliding scenarios, and dynamic collision surfaces. The results show that this method has the potential to successfully prevent collisions and allow haptic interaction for highly dynamic environments. We believe that this work significantly adds to the usability of current exoskeletons by enabling virtual haptic interaction with the patient’s body parts in human-robot therapy.

Publications 1 - 10 of 11

Michael Sommerhalder

Last Name

First Name

ORCID

Organisational unit

Filters

Search Results