Universität Stuttgart
Permanent URI for this communityhttps://elib.uni-stuttgart.de/handle/11682/1
Browse
21 results
Search Results
Item Open Access Engineered bioinspired natural dynamics and their synergy with control and learning in legged robots(2022) Ruppert, Felix; Schmitt, Syn (Prof. Dr.)The performance of legged locomotion relies on the successful mitigation of unstructured, rough terrain in the presence of sparse information and neurosensory delays. Bioinspired walking systems benefit from carefully engineered passive compliant behavior that models the inherent elastic behavior of muscle-tendon structures in animals. To leverage the passive behavior that provides energy efficiency, passive stability as well as simplified control and learning tasks to the system, locomotion control and learning algorithms have to be designed and coordinated with the natural system dynamics in mind to achieve similar locomotion behavior we see in animals. The major contribution of this thesis is the synergy of a bio-inspired leg design with biarticular muscle-tendon structures, a wearable force and pressure sensor design for closed-loop control in legged locomotion, a biologically inspired closed-loop central pattern generator with reflex-like feedback and a learning approach that enables the locomotion controller to leverage the carefully engineered natural dynamics of the robot to learn convincing locomotion skills and increase energy efficiency. The first contribution is a biologically inspired leg design focusing on the biarticular lower leg muscle-tendon structure in vertebrate animals. The biarticular elasticity provides two-dimensional passive impedance to the leg and allows the storage of energy orthogonal to the leg axis direction. The leg blueprint is characterized in its capability to store and release elastic energy in the biarticular structure. The stored energy can be recuperated back into the system and increases the energy efficiency of the leg. This leg design achieves the lowest relative cost of transport documented for all dynamically hopping and running robots. The second contribution introduces the concept of training wheels, temporary mechanical modifications to the system dynamics that shape the learning reward landscape and simplify learning locomotion directly in hardware. Through deliberate changes to the system dynamics, in this case, reduced gravity, the reward landscape can be shaped to simplify the learning process. Learning with this training wheel is safer due to smoother reward landscapes with shallower gradients. Also, the initial guess for initiating the machine learning algorithm is simplified, because the salient gradient set of viable reward signals is bigger. During the learning process, the training wheel influence can be gradually reduced and the learning algorithm converges to the solution of the initial learning problem without training wheels. The third contribution presents a rugged, lightweight force and pressure sensor for feedback information and biomechanical analysis. The sensor can be mounted on a robotic foot and provides continuous force and pressure feedback during locomotion in unstructured and soft terrain. The sensor is based on a pressure sensor, encapsulated in a polyurethane dome with an air cavity around the pressure sensor. External forces deform the sensor dome and the rising pressure in the air cavity is measured by the pressure sensor. Based on the dome geometry and material, the sensor range can be adjusted for different load cases. The sensor can be used in arrays to measure pressure distributions as well as a wearable force sensor in wet or granular media where classical force plates can not be utilized. The final contribution synergizes the individual contributions into a neuroinspired learning approach that matches a bioinspired closed-loop central pattern generator with reflex-like neuroelastic feedback to the natural dynamics of a quadruped robot with biarticular legs. Through sparse contact feedback from the foot sensor, the bioinspired central pattern generator can neuroelastically mitigate short-term perturbations to adapt the robot to its environment. Because the robot dynamics and the control task dynamics initially do not match, the controller uses the neuroelastic feedback to minimize the discrepancy between commanded and measured robot behavior. The amount of feedback activity during level walking can be used as a proxy to estimate the amount of dynamics mismatching. By minimizing the amount of required neuroelastic feedback the robot learns to neuroplastically match its control task dynamics to its natural dynamics through Bayesian optimization. Through the synergy of mechanics and control the biomechatronic system benefits from both the individual functionality of its components as well as their interplay during locomotion. The designed natural dynamics provide advantageous passive behavior to the robot and the bioinspired controller learns to leverage the natural dynamics to achieve convincing locomotion.Item Open Access Generation, probing, and biophysical stimulation of human microtissues in microfluidic Organ-on-Chip platforms(Stuttgart : Institute for Modelling and Simulation of Biomechanical Systems, Chair of Continuum Biomechanics and Mechanobiology, University of Stuttgart, 2022) Schneider, Oliver; Röhrle, Oliver (Prof., PhD)Over the last decade Organ-on-Chip (OoC) emerged as disruptive technology combining aspects of microfluidics and tissue engineering. OoCs culture human tissues in tailored microenvironments under microfluidic perfusion, yielding an unprecedented recapitulation of human physiology. So far, most systems predominantly focus on physiological tissue generation. However, it is crucial to integrate stimulation and readout capabilities, leveraging OoCs from bare tissue generation tools to advanced integrated experimental platforms. This thesis focuses on the development and characterization of novel microphysiological systems to probe and actuate tissues on the microscale. We present two Heart-on-Chip platforms enabling the generation of aligned cardiac muscle fibers and investigate the integration of force and O2 sensing as well as electrical stimulation capabilities. Furthermore, we introduce and characterize two OoCs enabling the precise delivery of biomechanical stretch and compression stimuli. All in all, the systems developed in the framework of this thesis provide a flexible toolkit amenable for disease modeling or personalized medicine, offering advanced experimental capabilities for manipulating and interrogating integrated tissues.Item Open Access Data-driven modelling of neuromechanical adaptation in skeletal muscles in response to isometric exercise(Stuttgart : Institute for Modelling and Simulation of Biomechanical Systems, Chair of Continuum Biomechanics and Mechanobiology, University of Stuttgart, 2022) Altan, Neriman Ekin; Röhrle, Oliver (Prof., PhD)This study aims to model the changes in the behaviour of motor neurons of the vastus lateralis in response to unilateral isometric knee extension exercise (UIKEE). For this, the phenomenological motor control model by Fuglevand et al. (1993) has been used. Input parameters for this model have been calibrated against data from experimental studies available in literature by using Bayesian updating. The pre-exercise state of the motor neuron pool of the muscle describing the recruitment behaviour as well as the contractile properties of the motor neurons have been constructed. Data collected from a systematic review on the change in isometric strength due to UIKEE has been modelled using Bayesian lonigutidinal model-based meta-analysis. Using the model of the change in isometric strength, increase in the average motor neuron discharge rate following UIKEE has been quantified.Item Open Access Development and implementation of next-generation Retina-on-Chip platforms(Stuttgart : Institute for Modelling and Simulation of Biomechanical Systems, Chair for Continuum Biomechanics and Mechanobiology, University of Stuttgart, 2024) Chuchuy, Johanna; Röhrle, Oliver (Prof., PhD)Item Open Access Insights into human alpha-motoneuron discharge properties during stretch reflexes : an in-silico approach(Stuttgart : Institute for Modelling and Simulation of Biomechanical Systems, Chair of Continuum Biomechanics and Mechanobiology, University of Stuttgart, 2024) Schmid, Laura; Röhrle, Oliver (Prof., PhD)Item Open Access Multilevel convergence analysis : parallel-in-time integration for fluid-structure interaction problems with applications in cardiac flow modeling(Stuttgart : Institute for Modelling and Simulation of Biomechanical Systems, Chair of Continuum Biomechanics and Mechanobiology, University of Stuttgart, 2020) Hessenthaler, Andreas; Röhrle, Oliver (Prof., PhD)In this Ph.D. Thesis, multigrid-reduction-in-time (MGRIT) is considered as means to reduce the time-to-solution for numerical algorithms concerned with the solution of time-dependent partial differential equations (PDEs) arising in the field of fluid-structure interaction (FSI) modeling. As a parallel-in-time integration method, the MGRIT algorithm significantly increases the potential for parallel speedup by employing modern computer architectures, ranging from small-scale clusters to massively parallel high-performance computing platforms. In this work, the MGRIT algorithm is considered as a true multilevel method that can exhibit optimal scaling. Convergence of MGRIT is studied for the solution of linear and nonlinear (systems of) PDEs: from single- to multiphysics applications relevant to FSI problems in two and three dimensions. A multilevel convergence framework for MGRIT is derived that establishes a priori upper bounds and approximate convergence factors for a variety of cycling strategies (e.g., V- and F-cycles), relaxation schemes and parameter settings. The convergence framework is applied to a number of test problems relevant to FSI modeling, both linear and nonlinear as well as parabolic and hyperbolic in nature. An MGRIT variant is further proposed that exploits the time-periodicity that is present in many biomedical engineering applications, e.g., cyclic blood flow in the human heart. The time-periodic MGRIT algorithm proves capable of consistently reducing the time-to-solution of an existing simulation model with significant observed speedups.Item Open Access Bioelectromagnetic fields for studying neuromuscular physiology : in silico investigations of EMG and MMG(Stuttgart : Institute for Modelling and Simulation of Biomechanical Systems, Chair of Continuum Biomechanics and Mechanobiology, University of Stuttgart, 2023) Klotz, Thomas; Röhrle, Oliver (Prof., PhD)Skeletal muscles generate bioelectromagnetic fields that contain information about the neural control of motions and the function of the muscle. One distinguishes between electromyography (EMG), the measurement of the muscle-induced electric potential field, and magnetomyography (MMG), the recording of muscle-induced magnetic fields. EMG is a well-established methodology, and its limitations have been extensively discussed in the scientific literature. In contrast, MMG is an emerging methodology with the potential to overcome some of the inherent limitations of EMG. To unlock the full potential of MMG, it is essential to support empirical observations from experiments with a solid theoretical understanding of muscle-induced bioelectromagnetic fields. Therefore, this thesis derives a novel multiscale skeletal muscle model that can predict realistic EMG and MMG signals. This model is used to conduct the first systematic comparison between surface EMG and non-invasive MMG. By using simulations, all system parameters can be controlled precisely. This would not be possible experimentally. The fundamental properties of EMG and MMG are systematically explored using simulations comparable to electrically or reflex-evoked contractions. Notably, it is shown that non-invasive MMG data is spatially more selective than comparable high-density EMG data. This property, for example, is advantageous for decomposing signals of voluntary contractions into individual motor unit spike trains. Using a novel in silico trial framework, it is demonstrated that non-invasive MMG-based motor unit decomposition is superior to the well-established surface EMG-based motor unit decomposition.Item Open Access The benefit of muscle-actuated systems : internal mechanics, optimization and learning(Stuttgart : Institut für Modellierung und Simulation Biomechanischer Systeme, Computational Biophysics and Biorobotics, 2023) Wochner, Isabell; Schmitt, Syn (Prof. Dr.)We are facing the challenge of an over-aging and overweight society. This leads to an increasing number of movement disorders and causes the loss of mobility and independence. To address this pressing issue, we need to develop new rehabilitation techniques and design innovative assistive devices. Achieving this goal requires a deeper understanding of the underlying mechanics that control muscle-actuated motion. However, despite extensive studies, the neural control of muscle-actuated motion remains poorly understood. While experiments are valuable and necessary tools to further our understanding, they are often limited by ethical and practical constraints. Therefore, simulating muscle-actuated motion has become increasingly important for testing hypotheses and bridge this knowledge gap. In silico, we can establish cause-effect relationships that are experimentally difficult or even impossible to measure. By changing morphological aspects of the underlying musculoskeletal structure or the neural control strategy itself, simulations are crucial in the quest for a deeper understanding of muscle-actuated motion. The insights gained from these simulations paves the way to develop new rehabilitation techniques, enhance pre-surgical planning, design better assistive devices and improve the performance of current robots. The primary objective of this dissertation is to study the intricate interplay between musculoskeletal dynamics, neural controller and the environment. To achieve this goal, a simulation framework has been developed as part of this thesis, enabling the modeling and control of muscle-actuated motion using both model-based and learning-based methods. By utilizing this framework, musculoskeletal models of the arm, head-neck complex and a simplified whole-body model are investigated in conjunction with various concepts of motor control. The main research questions of this thesis are therefore: 1. How does the neural control strategy select muscle activation patterns to generate the desired movement, and can we use this knowledge to design better assistive devices? 2. How does the musculoskeletal dynamics facilitate the neural control strategy in accomplishing this task of generating desired movements? To address these research questions, this thesis comprises a total of five journal and conference articles. More specifically, contributions I-III of this thesis focus on addressing the first research question which aims to understand how voluntary and reflexive movements can be predicted. First, we investigate various optimality principles using a musculoskeletal arm model to predict point-to-manifold reaching tasks. By using predictive simulations, we demonstrate how the arm would move towards a goal if, for example, our neural control strategy would minimize energy consumption. The main finding of this contribution shows that it is essential to include muscle dynamics and consider tasks with more openly defined targets to draw accurate conclusions about motor control. Through our analysis, we show that a combination of mechanical work, jerk and neuronal stimulation effort best predicts point-reaching when compared to human experiments. Second, we propose a novel method to optimize the design of exoskeleton power units taking into account the load cycle of predicted human movements. To achieve this goal, we employ a forward dynamic simulation of a generic musculoskeletal arm model, which is first scaled to represent different individuals. Next, we predict individual human motions and employ the predicted human torques to scale the electrical power units employing a novel scalability model. By considering the individual user needs and task demands, our approach achieves a lighter and more efficient design. In conclusion, our framework demonstrates the potential to improve the design of individual assistive devices. The third contribution focuses on predicting reflexive movements in response to sudden perturbations of the head-neck complex. To achieve this, we conducted experiments in which volunteers were placed on a table while supporting their heads with a trapdoor. This trapdoor was then suddenly released leading to a downward movement of the head until the reflexive reaction of the muscles stops the head from falling. We analyzed the results of these experiments, presenting characteristic parameters and highlighting differences between separate age and gender groups. Using this data, we also set up benchmark validations for a musculoskeletal head-neck model, including reflex control strategies. Our main findings are that there are large individual differences in reflexive responses between participants and that the perturbation direction significantly affects the reflexive response. Furthermore, we show that this data can be used as a benchmark test to validate musculoskeletal models and different muscle control strategies. While the first three contributions focus on the research question (1), contributions IV-V focus on (2) whether and how the musculoskeletal dynamics facilitate the learning and control task of various movements. We utilize a recently introduced information-theoretic approach called control effort to quantify the minimally required information to perform specific movements. By applying this concept, we can for example quantify how much biological muscles reduce the neuronal information load compared to technical DC-motors. We present a novel optimization algorithm to find this control effort and apply it to point-reaching and walking tasks. The main finding of this contribution is that the musculoskeletal dynamics reduce the control effort required for these movements compared to torque-driven systems. Finally, we hypothesize that the highly nonlinear muscle dynamics not only facilitate the control task but also provide inherent stability that is beneficial for learning from scratch. To test this, we employed various learning strategies for multiple anthropomorphic tasks, including point-reaching, ball-hitting, hopping, and squatting. The results of this investigation demonstrate that using muscle-like actuators improves the data-efficiency of the learning tasks. Additionally, including the muscle dynamics improves the robustness towards hyperparameters and allows for a better generalization towards unknown and unlearned perturbations. In summary, this thesis enhances existing methods to control and learn muscle-actuated motion, quantifies the control effort needed to perform certain movements and demonstrates that the inherent stability of the muscle dynamics facilitates the learning task. The models, control strategies, and experimental data presented in this work aid researchers in science and industry to improve their predictions in various fields such as neuroscience, ergonomics, rehabilitation, passive safety systems, and robotics. This allows us to reverse-engineer how we as humans control movement, uncovering the complex relationship between musculoskeletal dynamics and neural controller.Item Open Access Biophysical validity of reduced soft tissue modelling in neuro-musculoskeletal simulations(Stuttgart : Institut für Modellierung und Simulation Biomechanischer Systeme, Computational Biophysics and Biorobotics, 2024) Hammer, Maria; Schmitt, Syn (Prof. Dr. rer. nat.)In the past decades, neuro-musculoskeletal simulations have become a key technology in biomechanical research, and are increasingly utilised to support clinical decision-making processes, evaluate occupational safety, and facilitate the design of assistive devices. The importance of precise and physiologically valid simulations cannot be emphasised enough across all application fields. However, achieving this accuracy becomes particularly challenging when developing reduced descriptions of soft tissue compartments, where the degrees of freedom and structural complexity are condensed into predominantly phenomenological or homogenised sub-models. Furthermore, it necessitates a deep understanding of the dynamic interplay among different soft tissue elements in the body, which, in turn, requires a high level of reliability and confidence in the employed underlying model. In order to ensure the usefulness of the models, it is crucial to strike a balance between the level of detail and the limitations arising from simplifications. This involves considering various possibilities to validate the mechanical behaviour of each sub-model individually and the overall model as a whole. The scientific aim of this dissertation is to investigate the load sharing of soft tissue compartments at the example of the human lumbar spine during active motions by using predictive simulations. To set the basis for this kind of research, the main objective is to create, calibrate, verify and validate a detailed neuro-musculoskeletal model, which gives rise to three research questions that guide this thesis: (1) How can (and should) common approaches for reduced single tissue models be improved to increase the level of biomechanical validity and physical verification of both the sub-models themselves, and the multibody models composed of them? (2) Which sub-structures are of biomechanical relevance for estimating internal forces and torques on a full-body scale? (3) Which validation methods need to be considered during the development of a physiological spine model, and how can the corresponding simulation results be assessed? This thesis encompasses five journal articles studies, contributing to the different aspects of the three research questions. More precisely, Contribution I addresses research questions (1) and (3) regarding how to enhance biomechanical validity of muscle routing in multibody models by introducing a novel algorithm for redirecting muscle paths. With this method, the physiological accuracy of muscle length and moment arm representations within Hill-type models can be refined, particularly for muscles spanning multiple joints with multiple degrees of freedom. Contribution II focuses on the intuitive assessment of the relative motion between two vertebral bodies. Using a newly developed method for graphical representation of finite helical axes, which fully encapsulates the information about rotation and translation of the relative motion between two vertebral bodies, extensive data sets were effectively presented through clustering methods. This work, thereby, contributes to research question (3) since the finite helical axis can serve as measure for model validation. Contribution III presents a comprehensive simulation study and introduces a detailed generic model of the human thoracolumbar spine. This generic model includes several hundred muscle and ligament strands, along with intervertebral joints modelled as free joints constrained only by intervertebral discs. Notably, the model is able to balance gravity in an upright position without additional constraints and with only a physiologically low level of muscle stimulation. The description of model development and validation significantly contributes to research question (3). Moreover, the analysis of load sharing among the three soft tissue sub-structures, namely ligaments, muscles, and intervertebral discs, revealed an almost equal distribution of bending moment during forward flexion, offering insights for research question (2). Additionally, this paper introduces a workflow for geometric individualisation of the generic model based on landmark data obtained from computed-tomography scans. The investigation of subject-specific forces and torques exhibits significant inter- and intrapersonal differences in the lumbar load distribution. These findings deepen the biomechanical understanding of complex interactions within the spine. Contribution IV explores the influence of neglecting entire passive tissue groups, precisely intervertebral discs and ligaments, which is common practice in many spine models. Using an inverse dynamic approach, this study contributes to research questions (2) and (3) by providing cross-platform validation. The examination of kinematic models that exclude ligament and intervertebral disc tissues reveals a tendency for highly overestimated muscle forces. These findings highlight the importance of considering all relevant soft tissue structures in computational models to ensure accurate muscle force estimations. Lastly, the Contribution V directly addresses research question (1) by tackling the challenge of incorporating energy conservation in surrogate models representing the elastic mechanical responses of intervertebral discs. It introduces a novel approach that surpasses currently used elastic models in terms of precision, coupling of different degrees of freedom, and nonlinear behaviour. This advancement increases the accuracy and physical validity while also enabling the individualisation of the mechanical behaviour of subject- and level-specific intervertebral disc geometry. In order to achieve the aforementioned aims and answer the research questions, a multibody simulation framework has been extended by the novel algorithms developed within the scope of this thesis. These algorithms improve the geometric quality of soft tissue representations and refine accuracy of predicted internal forces and torques while preserving basic physical principles. Furthermore, a pre-processor typically used to scale population-based models has been modified to serve two additional purposes. On the one hand, the existing code was advanced to create geometrically individualised models based on landmark positions derived from computed-tomography scans. On the other hand, the functionality to generate models compatible with another widely used multibody simulation tool was included. Having identical models facilitates cross-platform validation, and allows to test muscle, ligament and intervertebral disc sub-models, and whole thoracolumbar spine models in different scenarios. In summary, this thesis enhances existing methods and provides new approaches to create more accurate neuro-musculoskeletal models capable of predicting internal forces and kinematics. The detailed spine model developed during this thesis serves as a valuable foundation for future investigations, including subject-specific, non-invasive implant testing, and exploration of the rotation axes in complex movements and post-surgical scenarios.Item Open Access On the load limits of the muscle-tendon unit and their applications in musculoskeletal human body models(Stuttgart : Institut für Modellierung und Simulation Biomechanischer Systeme, Computational Biophysics and Biorobotics, 2024) Nölle, Lennart V.; Schmitt, Syn (Prof. Dr.)The human skeletal muscle fulfils many movement-related functions, simultaneously acting as the main motor, spring, strut and brake of the body. Equally important for human motion generation are the tendons, which provide passive joint stabilisation and transfer the muscle’s contraction forces to the skeletal structure. Together, muscle and tendon form the muscle-tendon unit (MTU). Despite its ability to withstand many different loading scenarios, the MTU is susceptible to numerous kinds of injury, the most prevalent being the muscle strain injury. The retrospective evaluation of observed injury scenarios and the prediction of injury outcomes and risks has been increasingly important in sports biomechanics, automotive safety and forensic traumatology. For this purpose, numerous injury criteria have been defined for the use with both physical and virtual representations of the human body. While significant efforts in the field of injury severity classification have been made, strain injuries of the MTU have not yet been taken into consideration. This might be because conventional methods of defining injury criteria are not applicable to MTU strain injuries as the properties of the MTU and the nature of MTU strain injuries pose numerous unresolved challenges so far. The primary objective of this dissertation is to overcome these challenges and to define and substantiate MTU strain injury criteria for the use in musculoskeletal human body model simulations. The overarching research question which the presented thesis aims to answer is how injury criteria for strain injuries of the MTU can be defined and which information can be derived from their application. Throughout, the following sub-questions are addressed: 1. How can a strain injury criterion for the muscle be defined and substantiated based on literature data? 2. How can a strain injury criterion for the tendon be defined and applied to the recreation of an injury load case? 3. Which other applications besides injury severity assessment exist for the proposed injury criteria? These questions were tackled consecutively in three journal publications which comprise this dissertation. Sub-question 1 was answered in Contribution 1, where a muscle strain injury criterion (MSIC) was defined based on experimental data from the literature. The resulting injury criterion can differentiate between three levels of injury severity and is easily applicable to the computational representation of any muscle. The injury thresholds were substantiated by comparison to the calculated maximum ultimate tensile strength of mammalian skeletal muscle and through the application of the MSIC in a sprinting gait cycle simulation. The MSIC was also used for a simulation study on the aetiology of muscle strain injuries in which it was shown that material inhomogeneities might cause localised strain injuries within a muscle. To tackle sub-question 2, Contribution 2 built on the findings of Contribution 1 by formulating the tendon strain injury criterion TSIC. This criterion was used to investigate the forces and strains acting on finger flexor tendons during jersey finger injury scenarios. For this purpose, a finite element neuromusculoskeletal hand model was created through the combination of two preexisting models. Additionally, new Hill-type muscle elements were inserted whose parameters were calibrated to fit experimental data. The newly created hand model was used to recreate a simplified jersey finger injury load case under varying muscle activity levels. This simulations study showed that a correlation between muscle activity and sustained injury severity exists. Finally, Contribution 3 set out to answer sub-question 3 and to demonstrate the usefulness of the MSIC and TSIC for applications other than injury severity assessment. For this, common modelling issues present in musculoskeletal human body models were first recreated and then detected using the proposed criteria. First, the deformation of a finite element model’s skeletal structure during model repositioning was identified through an MSIC assessment of muscles spanning a displaced joint. Second, an ill-tuned muscle parameter within an otherwise physiological model was found through applying the TSIC to a multibody gait cycle simulation. Additionally, a new method for determining minor TSIC thresholds for arbitrarily parameterised tendons was developed, thus improving the usability of the TSIC. The cumulative result of this thesis is a strain injury criterion for the MTU which, to the author’s knowledge, is the first of its kind. Additionally, a new method for evaluating the quality of musculoskeletal human body models was provided. Future studies should focus on the experimental validation of the proposed injury criteria and on expanding them by statistical metrics. Potential application scenarios of the MSIC and TSIC, besides injury evaluation, are as model assessment tools or in ergonomics.
- «
- 1 (current)
- 2
- 3
- »