The autonomous inspection of subsea infrastructure is significantly hindered by marine biofouling, which frequently causes traditional rigid crawlers to stall or necessitates expensive, time-consuming surface cleaning. Furthermore, the subsea infrastructure itself introduces significant obstacles, as the path is frequently interrupted by geometrically complex architectures such as interconnected valve assemblies, flanges or varying pipeline diameters. To address these limitations, this research presents the design, computational modelling, and empirical validation of the Compliant Robotic Architecture for Biofouling (CRAB) prototype. The CRAB leverages inherent material compliance to overcome these obstacles on pipelines.

The architecture integrates three core subsystems: a tendon-driven, underactuated gripper for adaptive enclosure, a passive magnetic sliding track for variable circumference locking, and fluid-filled flexible wheels designed to deform over obstacles. Finite element analysis was
utilised to optimise the wheel morphology.

Empirical validation of the prototype confirmed the viability of the core design concept, with the CRAB successfully achieving a secure grasp and overcoming simulated radial biofouling up to 50 mm in radius. However, testing also exposed critical failures, specifically material ruptures at 60 mm obstacle and kinematic stalling within the variable locking mechanism.

Ultimately, this research validates the foundational methodology of using passive mechanical compliance for unstructured subsea mobility. While the core kinematics are proven effective, advancing the system toward autonomous field deployment requires the integration of anactive mobility actuation system, comprehensive dynamic stability analysis, structural refinements and material optimisation. ... Read more

Conventional articulated robot arms excel on structured production lines but remain unsafe and ineffective in cluttered, dynamic settings, while existing soft-robotic manipulators sacrifice load-bearing capacity and positional repeatability for safety and adaptability, origami-robots improve this trade-off. The octopus arm, a muscular hydrostat, is a biological example of overcoming this trade-off by uniting shape morphing, distributed actuation and tunable stiffness. This thesis introduces a fold-flat, modular origami structure that co-locates distributed magnetic actuation and thermally tunable stiffness inside each unit module, thereby matching the octopus arm’s three biological features, shape morphing (1), distributed actuation (2) and tunable stiffness (3). A four-legged water-bomb module was created by mapping biologically derived requirements onto an additively manufactured construction that integrates paired NdFeB permanent magnets, a central pancake electromagnet, and a Joule-heated conductive-PLA crease. Physical prototypes and finite-element analysis predicted ±20 ° biaxial bending, 100 % axial extension, and a 50% stiffness reduction. A proof-of-concept prototype verified these predictions, except that the bending was up to 10°. Single-leg or single-hinge tests revealed a steep modulus drop between 45 °C and 60°C, yielding up to 90 % stiffness reduction under 34 V excitation. At module level the actuator stack produced 10 mm axial stroke, ±10 ° bending and peak push/pull forces of 0.8 N/1.0 N while maintaining repeatable 3-DoF motion. Eight of the ten primary requirements were met; sustained horizontal self-support and whole-arm 90 ° curvature were limited by cable weight and hinge play. The study identified two bottlenecks: high coil currents that drive the PLA above its glass transition, and asymmetric flux coupling when the coil drifts toward one magnet. Improvement strategies include closed-loop thermal control, heat-resistant substrate polymers, and improved electromagnet positioning. By uniting shape morphing, distributed actuation, and tunable stiffness in a centimetre-scale, magnetically actuated origami cell, this work realises an octopus-like robotic arm with power-off stiffening, planar manufacturability, and modularity. The results establish a viable route toward deployable continuum arms for confined-space inspection, human-robot collaboration, and search-and-rescue operations. ... Read more

Congenital heart disease (CHD) affects almost 1% of newborns. Right ventricular outflow tract (RVOT) CHD affects 20% of newborns and includes anomalies such as tetralogy of Fallot (TOF) with or without pulmonary atresia, transposition of the great vessels, and truncus arteriosus. All these anomalies require RVOT reconstruction. Prosthetic heart valves are needed to improve the quality of life of patients suffering from CHD. One such prosthetic device is the pulmonary valved Conduit developed by XeltisTM. This study aimed to investigate the difference in the mechanical response of the XeltisTM pulmonary valved Conduit sizes 16, 18, and 20 (XPV16, XPV18, and XPV20) using mechanical experiments and a predictive finite element model.
Experiments of two load cases, Leaflet opening behavior (LC1) and parallel compression (LC2) have been done where measurements were taken for input parameters used for uncertainty quantification (UQ) in the FE model. Furthermore, the reaction force and displacement were measured to calculate the force values and stiffness values of the device during each experiment. The experiments were replicated with a developed FE model. From the results of the FE model, a metamodel (MM) was developed and a Monte Carlo simulation was performed to retrieve a distribution of the force values and stiffness values obtained in the simulation. Furthermore, UQ was performed and the sensitivity of the input parameters on the force values and stiffness values were quantified. Finally, with an area metric the accuracy of computational finite element (FE) models in simulating the mechanical response observed in the experiments of the XeltisTM pulmonary valved Conduit size 16, 18, and 20 mm was quantified.
For the Leaflet opening behavior, the reaction force increases when the device size increases while the stiffness doesn’t change with device size. As for the accuracy of the predictive FE model, the predictive FE model can simulate the mechanical response of the stiffness with an accuracy of at least 70.7% for the reaction force and at least 50.3% for the stiffness.
For the parallel plate compression, the reaction force and the stiffness increase when the XPV size decreases from size 18 to size 16. Furthermore, the difference between the XPV18 and the XPV20 is smaller for both the reaction force and the stiffness. As for the accuracy of the predictive FE model, the predictive FE model can simulate the mechanical response of the stiffness with an accuracy of at least 37.3% for the reaction force and at least 38.1% for the stiffness.
As for the important variables influencing the mechanical response, the reaction force and the stiffness are influenced mostly by the fiber stiffness of the component that is subjected to the load. Furthermore, the reaction force and stiffness are also influenced by the direction of the fibers. If the fibers are in the same direction as the load, the reaction force and stiffness increase. Although specifically for the Leaflet the amount of material has more influence on the reaction force and stiffness for larger XPV sizes with the Leaflet opening behavior load case. This indicates that for larger XPV sizes the material properties of the Leaflet have a smaller influence and the Leaflet geometry has a higher influence on the stiffness of the Leaflet opening.
Improvements are possible for a better agreement between the simulation and the experiment. These include performing more experiments with different samples from different production batches, and better estimation of input parameters with a high sensitivity to the output parameters.
This study provides a step in the direction of predictive computational device modeling that will help shorten the development time of new pulmonary heart valve devices. As the devices in this study are designed for pediatrics, this will help improve the quality of life of pediatrics suffering from congenital heart disease. ... Read more

Intro - The rapidly developing technology of cardiac finite element modelling aims to improve heart failure treatment by quantifying stresses acting in the cardiac tissue. Cardiac finite element modelling may improve heart failure treatment by providing more insight in the pathophysiology, enabling a patient-specific assessment and improving the efficiency of clinical trials and medical devices. An essential part of the cardiac finite element model is capturing the mechanical behaviour of the heart, including its active behaviour (contraction). This mechanical behaviour is captured by means of a material model. However, the active behaviour of the heart is not readily available in material models provided by commercial finite element software like Abaqus/CAE. Consequently, such active material model needs to be added in a user subroutine manually. This is a time-consuming task, prone to errors. Additionally, the availability of the documentation is limited, and in literature the methodology of the implementation is oftentimes not disclosed.

Thesis goal - The goal of this master-thesis is to implement a combination of the passive and active mathematical material models in a commercial finite element platform. This was done by implementing the combined material models in user subroutine UMAT that can be used in FEA-software Abaqus. Reproducibility of the UMAT is ensured by a detailed documentation. Also, the UMAT is provided in the supplementary material.

Methods - The UMAT consists of the Holzapfel Ogden constitutive law as passive component, and the Time Varying-Elastance contstitutive law as active component, which are combined by means of the active stress approach. Additionally, the UMAT requires computation of the elasticity tensor, which is computed by means of a numerical formulation. The incorporation of the passive and active constitutive laws in the UMAT was verified by means of multiple test-cases. The outcomes of the UMAT were compared to an analytical solution and a benchmark user subroutine in the form of UANISOHYPER_INV. Lastly, reality-check test-cases were carried out by comparing the UMAT outcomes to the results from similar test-cases found in literature.

Verification results - Verification of the implemented material models showed good agreement with the analytical computed solution of equibiaxial extension, equibiaxial compression and isometric contraction test-cases, as all cases showed an MAPEmax or APEmax error lower than 1%. Shear test-case results of the UMAT showed some bigger APEmax values (maximal 17%) with relation to the analytical solution, possibly caused by numerical errors during the elasticity tensor computation. Results of the reality-check cases showed similar trends to the mechanical experiments done on cardiac tissue on which these cases are based.

Conclusion - The public availability of the implemented passive and active material models in the user subroutine UMAT, which is working reasonably well according to conducted verification, forms a significant step forwards in the field of cardiac finite element modelling. The current work can be further extended by the incorporations of compressibility during contraction and viscoelasticity in the material models. This thesis provides the basis for future projects in the field of cardiac finite element modelling including an active material model. Also, the provided implementation of material models may aid in the implementation of other mathematical material models in a user subroutine like UMAT. ... Read more

Atherosclerotic plaque rupture is the underlying cause of 50% of deaths in western society. Although screening methodologies exist, plaques are highly complex, and the parameters used to measure plaque vulnerability are often insufficient for correct medical screening. Since plaque rupture occurs when the vascular forces exceed the plaque strength, the mechanical analysis of plaque stability could offer a new window into predicting rupture. By interpreting the mechanical behaviour of plaques, so-called mechanical markers could be derived, which could highlight vulnerable plaques before thrombosis. Unfortunately, the extensively modelled local stresses and energy functions are immeasurable in vivo. Therefore we turned to strains, a measure of material deformation that can be obtained clinically. This study investigates the predictive value of strain distributions in the early detection of fibrous caps rupture.
Simplistic plaque cap mimics were engineered to model atherosclerotic plaque rupture. Made from a fibrinous matrix and a soft lipidic inclusion (SI), the constructs were tailored to have mechanical properties similar to in vivo plaque caps. Tissue engineered caps offer a wide range of advantages over endarterectomy samples, including unlimited sample availability, robust geometry, high reproducibility and precise control over biological constituents. These constructs were uniaxially strained and subjected to 2D digital image correlation (DIC) to obtain their strain fields. Two-dimensional DIC is an algorithm that derives the material deformation by tracking surface features of its target sample. Therefore, it can very accurately calculate local strains for a material. This report analyses the patterns and maxima of the strain maps of five samples to evaluate unique features distinct at the rupture location that could serve as potential markers of plaque vulnerability. Besides inspecting the individual strain maps, their collective contribution through two strain-based failure criteria was analysed. These failure criteria approximate the strain energy in the samples.
The mechanical failure of the constructs was not instantaneous. All samples underwent failure in phases, starting with a small crack in the SI and concluding with the rupture of the fibrous tissue. Likewise, the strain and failure criteria patterns were consistent within all samples. Even though their strain values differ, the accumulations of low and high strains lie at nearly identical positions. These patterns emerged early in the tensile experiment, as the frames at a physiological (10%) and final (ultimate state before rupture) global strain measured a strong resemblance. In a more localised analysis, the fields were radially divided into ‘slices’ of data to produce distinct segment-based patterns. The rupture location consistently lies at a unique feature in the pattern, such as a peak or a valley. As was observed before, aside from the difference in their magnitude, the strain patterns showed no change between the frames. Finally, the local maxima of the strain and failure criteria maps were inspected. Their distances to the rupture site were comparable for all maps, indicating they performed similarly at estimating the rupture location. Moreover, the distances measured for the final frame showed no significant difference to those measured for the physiological frame.
The mechanical response of the five samples is similar. Aside from undergoing mechanical rupture in stages, the analysis of the tissue construct shows that a relationship exists between the local strains and rupture location. First, the rupture location always sits at the edge of a high valued region in the SI. Second, the tissue segmentation produces a highly reproducible pattern with a distinct colocalisation between segment patterns and the rupture location. Third, the maxima of the strain and failure criteria maps lie close to the rupture site, although they do not overlap. Accordingly, there is a clear relationship between the rupture location and the
local strain patterns. ... Read more