Scaling down robots is a challenge that requires innovative design approaches. Current locomotion robots are often dependent on complex actuation schemes or multiple actuators, which limits their miniaturization. Integrating control of a mechanism into the body instead of using complex electrical control architectures can contribute greatly to the scalability of locomotion robots. This research proposes a compliant mechanism that converts a rectilinear input into a closed-loop End Effector (EE) trajectory. Six of these mechanisms are connected through a compliant transmission to form a walking hexapod with a tripod gait. The result is a robot where a single linear actuator creates a walking gait with zero internal friction. This research hereby paves the way towards scalable locomotion robots with a simple control architecture.

Diamagnetic levitation provides a frictionless and entirely passive method of actuation, operating even in vacuum and at cryogenic temperatures. It has been explored in fields ranging from seismology to microfluidics handling, and its qualities make it attractive for MEMS devices by reducing production costs and improving robustness. This work develops a diamagnetic motion stage aimed at extending the achievable range of motion beyond prior nanometer-scale demonstrations. This objective is formulated as the following research question: What would be an optimal design for a long range actuator to create maximum move- ment range using diamagnetic levitation to eliminate contact friction and electrostatic force as propulsion method using only feedforward control, while keeping the cost low?. Magnet arrays were modeled, with a railroad configuration selected for its levitation properties. An electrostatic actua- tion model was derived, showing that force depends on both electrode area and its change as a result of movement, and that stable control requires balancing charged and grounded electrodes. Two systems were implemented: a low-voltage (0–200 V) setup with high control but low force, and a high-voltage (0–1200 V) setup with greater force but less control. Characterization revealed Duffing- type nonlinear dynamics. The low-voltage stage enabled three-phase frequency control, useful for resonance testing, while the high-voltage stage achieved up to 5.5 mm controlled travel and 23.47 mm one-way displacement, though with the end position being unrecoverable using electrostatics. The results show that longer travel requires repeating magnet arrays where all potential local minima support levitation. Another important finding was that electrostatic force has both x and z components, with their ratio depending on levitation height and stage surface area, highlighting the importance of thin electrodes. Scaling analysis suggests further potential for MEMS integration.

The precise and damage-free manipulation of fragile chips such as sensors, MEMS devices, or other micro-fabricated components with surface-bound structures demands grippers that can operate reliably within highly constrained assembly environments. These chips often feature delicate surface components that preclude traditional gripping strategies, especially in contexts where clearances are limited to sub-millimeter levels. This work develops and experimentally validates a generalizable framework called the Grasp Analysis Methodology (GAM) that combines Grasp Wrench Space (GWS) analysis with classical force distribution theory to support the design and evaluation of grippers for such applications. The framework is applied to three distinct gripper configurations: an industry-standard surface grasp gripper, a three-point gripper derived from the decomposition of a typical edge-based industrial gripper, and a four-point corner-supported design. Grasp quality is assessed using GWS metrics, which characterize each configuration’s ability to resist external disturbances through its wrench space properties. Concurrently, the mechanical loading experienced by the chip during grasping is estimated using plate theory to understand stress concentrations and risk of damage. To validate these analytical results and hence the gripper designs, a series of experimental tests is conducted. Pull tests evaluate disturbance resistance, while placement trials examine the gripper’s ability to preserve the chip’s orientation from pick up to placement, a critical requirement in precision assembly tasks. Unlike most academic gripper designs, which are either un-validated or unconstrained, this study emphasizes real world feasibility by testing under a spatial clearance of just 0.5mm. The findings demonstrate that the proposed GWS-Force strategy under the developed GAM workflow offers a robust basis for evaluating and optimizing chip grippers in constrained, high-precision environments

This thesis studies how to estimate cubic non-linear stiffness in dynamic systems using two-tone excitation. When a structure with cubic stiffness is driven at two frequencies that lie close to each other, it produces inter modulation peaks at sum and difference frequencies. By relating the amplitudes at these frequencies to the system parameters, the cubic stiffness can in principle be extracted without full-scale numerical modelling or fits with convoluted functions. An analytic relation between the amplitude at an excitation tone (A1) and the first-order sum frequency amplitude (A3) is derived using the harmonic balancing method, expressed in terms of the resonance frequency ω0, detuning δ, and cubic stiffness γ. The derivation uses a Ansatz with the corresponding sum frequencies and orders terms to obtain a analytic relation. This analytic curve is then validated in two ways: (i) by numerically solving the harmonic-balance equations, and (ii) by direct time-domain integration (RK45) with amplitude ex traction at the relevant frequencies. Symbolic regression is also used to fit compact additional formulas where the analytic relation loses accuracy. Experiments are carried out on Silicon-Nitride beams using a Polytec MSA-500 LDV and a Moku: Lab for signal generation and spectrum analysis. The main result is that the analytic A1–A3 relation agrees well with harmonic-balance and numeric integration simulations, but only within a certain region: low to moderate drive and sufficiently separated tones so that higher-order products remain small. Outside this region, solution branching and neglected terms reduce accuracy. In measurements, the expected near-linear relation between the two drive amplitudes (A1, A2) is observed, but the measured A1–A3 curves rise more steeply than predicted. This could be caused by additional effects such as higher-order non-linearities, damping, or resonance shifts under strong drive. The work delivers: (1) an analytic and numerical framework that maps where two-tone inter modulation can identify cubic stiffness, and (2) a robust experimental workflow. It also outlines future improvements: adding (non-linear) damping and quadratic stiffness to the model, lock-in based detection, better input conditioning, and resonance tracking to improve the differences between theory and experiments.

Many robotic applications require moving an end-effector through intricate closed-loop paths for object manipulation or locomotion. Conventionally, rotary-actuated rigid-link mechanisms perform this task successfully. However, several drawbacks, such as wear, play, and assembly difficulties, limit their performance. In high-speed applications, these rigid and often bulky links require high accelerations, leading to high power usage or the need for dynamic balancing, further complicating the mechanism. Rigid mechanisms are not the only ones that generate paths; compliant mechanisms are also widely used. However, compliant hinges cannot undergo complete rotations by definition, making cyclic actuation impossible. As a result, creating closed-loop paths typically requires multiple actuators—one per end-effector degree of freedom—adding to power demands and complexity. In contrast, closed-loop deformations can occur when dynamically actuating a soft body with a single actuator. We can create customizable soft bodies that leverage internal dynamics by structuring this soft body with a mechanical metamaterial made of tessellated compliant cells and designing its internal geometry. This research explores how these dynamics can be harnessed for path generation within mechanical metamaterials. Through multi-objective optimization, we embed reprogrammable control strategies within the metamaterial geometry, enabling adaptive responses to actuation frequency and amplitude for complex behaviors. We validate these designs with tabletop prototypes, building up to a self-propelled, walking prototype—a step toward autonomous robotic metamaterials.

Electric and hybrid vehicles are a common sight on the roads these days. The electric machine is known to be quiet, but sometimes high-frequency sounds are observed when driving. Over the years many different solutions have been developed to make the electric machine quieter, but almost no research looks into the possibility to generate sounds. This thesis looks into the possibility to produce nice sounds and reduce annoying sounds with electric machines. This is done by introducing harmonics into the current and creating additional excitations of the structure. To gain a complete understanding of this topic the electromagnetic domain and the structural domain have been analysed. A mathematical description of the forces has been created to be able to make predictions about the forces created by an injected current harmonic. Using the electromagnetic model the exact influence of current harmonics on the forces has been verified. The sound power level produced by the forces acting on the structure is calculated using the structural model. By properly selecting the current properties it is possible to create new temporal orders that produce sounds. For the analysed electric machine the sound power level of the desired sound is on average approximately 20dB lower than the sound power level of the fundamental current. Furthermore the sound power level is dominated by the excitation of the breathing modes. Using current injection existing sounds can be reduced. Using brute force optimisation the average radial forces were reduced. This reduces the sound power level during most of the run-up and reduces the influence of the breathing mode by up to 11dB. At some points during the run-up more sound is produced, because the tangential forces are increased, while the radial forces are decreased. In the end it is shown that the influence of the introduced current harmonics on the acoustics can be analysed using mathematics and the simulation results. New sounds can be produced, but the sound power level is limited by structural design of the electric machine.

Diamagnetic levitation presents a promising platform for realizing resonant sensors and energy harvesters. The technique offers mechanical isolation from the environment while operating with zero power consumption. This unique feature ensures exceptional sensitivity and accuracy in numerous applications. However, the presence of eddy currents in the levitating plate, induced by the alternating magnetic field, poses challenges, leading to increased damping and subsequently limiting the performance of levitating resonators. To address these issues, this study proposes a novel solution through the segmentation of diamagnetic plates. By dividing a pyrolytic graphite plate into smaller blocks, the flow of eddy currents is effectively restricted, resulting in reduced damping and significantly higher Q factors. By comparing the theoretical predictions using a FEM model from an earlier study with the measurements conducted the proposed technique is further validated. This comparative analysis demonstrates the effectiveness of the segmentation approach in mitigating damping due to eddy currents. Furthermore, the implementation of the proposed method led to remarkable results, with achieved Q factors exceeding 150 thousand.

Vibration attenuation in lightly damped blade-like systems such as cantilevers or end-effectors of robot manipulators can be dampened using Active Vibration Control (AVC). Piezoelectric patches in a collocated setup measure and control the bending modes of such a cantilever system using Positive Position Feedback (PPF). Tuning methods for PPF are based on optimisations for maximum disturbance rejection and disregard the presence of electronic noise. However, because electronic noise is amplified by controllers with high gains, it can become a significant error source in the system. In this paper, the effect of noise amplification is investigated for blade-like AVC systems. It is shown that there is a trade-off in the total dynamic error between vibration attenuation and noise amplification. An optimisation of this trade-off is proposed, which is performed on an industrial example of a blade-like system and validated experimentally. The optimisation shows clear improvement over traditional tuning methods that disregard the presence and effect of noise.

From the motion of electrons in an atom to the orbits of celestial bodies in the cosmos, governing equations are essential to the characterisation of dynamical systems. They facilitate an understanding of the physics of a system, which enables the development of useful techniques such as predictive control. An increasingly popular method to obtain these equations is Symbolic Regression, where governing laws are discovered from observations of the system. In this work, we extend the Deep Symbolic Regression package to the identification of dynamical systems. Preliminary tests revealed the limits of the method as applied to dynamical systems, and new methods of incorporating domain knowledge to constrain the expression space are added. We test the extended package on 3 strongly nonlinear ODEs that exhibit different dynamics as their parameterisation varies. Finally, we demonstrate the working of this method in practice by discovering the governing equation of a pendulum, from a video of its oscillation. This method achieved a 100% equation recovery rate on our tests, and was able to consistently retrieve the correct equation from datasets representing a diverse range of dynamics.

Vibration energy harvesting is a growing field of research. Harvesting energy from vibrations can be of use in devices in hard to reach places, such as sensors on a train track or a pacemaker. These devices makes use of batteries, which need to be replaced. Using energy harvesting this battery life could be improved.
The difficulty lies in harvesting low frequency vibrations. These vibrations are hard to harvest using a transducer such as a piezo. A solution could be using a frequency up converter, which can raise a vibrations frequency. This thesis shows the design of a Bi-stable Impact-driven Snap-through frequency up converter (BISup). Enhancing the behaviour of this bi-stable design could improve the energy harvesting capabilities. Different designs are made using ortho planar spring design, these designs are simulated and compared using ANSYS. These simulations are experimentally verified.

High-Speed Atomic Force Microscopy is widely used for investigating biological architectures and the semiconductor industry. The main limitation comes from parachuting or the Wile E. Coyote effect. Parachuting is the phenomenon where the cantilever taps on the sample towards a steep decrease in height, but does not get adjusted for this decrease fast enough, resulting in poor imaging quality at higher scan speeds. A proof-of-concept has been modeled for an adaptive raster scanning methodology on the X and Y piezo actuators. There will be investigated how a variable scan speed can decrease the effect of parachuting and thereby improve the spatial resolution. A detection algorithm is designed to measure the parachuting and uphill events at high scanning speed as accurately and precisely as possible during the forward scan. This has been done by defining a mixed signal that multiplies the first and second derivatives of the filtered deflection signal. A variable scan speed will be applied for the backward scan. The detection algorithm turned out to have a very high repeatability of 97.1% for the uphill events, and 93.6% for the parachuting events. Its accuracy turned out to have a maximum deviation of one signal period, which has been accounted for within the controller. Implementing this for the adaptive controller results in an improvement in both resolution and time efficiency. The adaptive controller is up to 9.5 times more accurate and time efficient compared to conventional methods.

Current service and installation activities to offshore wind turbines (OWTs) are mainly carried out by jack-up vessels. These vessels are less susceptive to disturbances caused by sea waves when they lift their hull from the water level by jacking down multiple legs to the seabed. An alternative is a crane mounted on a floating vessel, which has the benefits of being faster regarding sailing time, cheaper in operation, able to locate multiple times around the same offshore site and able to operate in deep waters. However floating crane vessels have a significant lower workability in comparison to jack-up vessels and are therefore not deployed in OWT activities yet.
Huisman initiated the three-dimensional (3D) motion compensated crane (MCC) project to improve the workability of crane vessels in offshore operations - thereby focusing on the performance of the equipped cranes - such that floating crane vessels will be able to install and service OWTs during an sufficiently large operational window. But to date MCCs have never been realized for offshore wind industry, which makes it difficult to estimate their performance beforehand and evaluate whether they are feasibile.
This thesis describes the development of a procedure to assess the feasibility of motion systems from a mechatronic design perspective and the motion compensation system of the Huisman 3D MCC is used as a case study. Several aspects of the MCC have been studied into more detail because these aspects are an determining factor regarding feasibility. These aspects are the influence of actuator type and crane boom flexibility and the performance during realistic offshore conditions. To investigate these aspects the system dynamics of the conceptual design are modelled, a controller is designed for the model and parameters are identified. Next the model is evaluated on its disturbance rejection capability because this is an important performance indicator for motion compensation, its robustness is examined and whether the OWT installation requirements are fulfilled.

Over the past decades the diamagnetic effect has been used for a variety of applications. This ranges from levitating a living frog to stabilizing a force sensor, designing an acceleration sensor and determin­ing the power of a laser. Diamagnetic levitation is the only method in which a passive stable levitation of an object can be achieved. Such a stable levitation has the benefits that the levitated object is iso­lated from its environment, thus eliminating mechanical damping and friction. Furthermore, no energy or control mechanisms are needed to sustain the levitation and to keep the object in its place. With the decrease in the damping, due to the elimination of the mechanical damping component, potential non­linearities in the movement of the plate can present themselves. Such nonlinearities can be exploited in, for example, energy harvesters. In such energy harvesters, the nonlinearities can be used to increase the range of frequencies at which energy can be harvested. As both diamagnetism and nonlineari­ties have their benefits, this study aims at investigating the nonlinear behaviour of a diamagnetically levitating plate in the vertical direction. An analytical model has been constructed which aimed at pre­dicting the (non)linear behaviour of a pyrolytic graphite plate, levitating over a 2x2 array of permanent magnets, which is actuated electrostatically. The model is verified numerically using COMSOL and has been used to predict both the static -­ and dynamic behaviour of the plate. Experiments have been conducted to assess the validity of the analytical predictions. It is found that the static behaviour and the linear dynamic behaviour of the plate can be predicted analytically within certain margins. The ana­lytical model indicates that all combinations of the magnetic force and electrostatic forces investigated have a softening effect on the levitating plate. A softening response is also observed in experiments. However, the analytical model cannot perfectly predict the experimental frequency response functions. A possible reasons for this discrepancy might be mode coupling. Overall, the study of the nonlinear dynamics for a diamagnetically levitating plate will help to gain more understanding on its fundamental mechanism and thus pave a way for its wider applications.

Any resonance based sensing method such as AFM or mass sensing using microfluidic cantilevers require high frequency stability. For mass measurement, a stable frequency allows for a lower mass resolution. One of the most commonly used methods to actuate these resonators is using piezoacoustics. A major downside of this method is the presence of spurious resonances, which corrugate the frequency response and can potentially degrade mass sensitivity. One way get rid of these spurious peaks is by
actuating photothermally (using laser light). This thesis is focused around designing and building a photothermal AFM to explore this. Additionally, some experiments are performed comparing piezoacoustic actuation to photothermal actuation in terms of the aforementioned frequency stability and
spurious peaks.

Vibration energy harvesting has proven to be a durable source of energy for a wide variety of applications, however not all of the positions these applications are placed at are suitable for conventional vibration energy harvesting. For some of these applications a frequency up-converted energy harvester could increase the power harvested. These systems are analysed in this thesis project, by first gaining insight in previously performed. After which a new method is designed and modelled. The design was then fabricated and a proof of concept was tested, of which the results are presented in this thesis project report.

It is relevant for various applications to seek a material that could be used for the active control of wave propagation. In this study, we explore tunablemetamaterials based on foldable origami sheets where the size of the bandgap can be tuned by the rigid folding along predefined crease lines. A combination of out-of-plane components and a crease results in the coupling and separation of bands in the band structure and eventually in a (modal) bandgap. The location, size, and vibration isolation performance are optimized and analyzed for a set of geometrical and material parameters, resulting in a potential design of aMiura-Ori structure. The methods used in this paper could be used for any kind of origami structure, which may lead to various new applications in the field of vibrations and acoustics.

With the incline of multidrug resistant bacteria due to the widespread use of antibiotics, there has been an ongoing search for a faster and more effective method to test the effect of antibiotics on bacteria. With current techniques for investigation of bacterial resistance requiring a minimum of 24 hours to complete, a new technique is essential to reduce the widespread misuse of antibiotics worldwide. In the last decade, new methods of this so-called Antibiotic Susceptibility Testing have been on the rise, with an emphasis on the use of nanomechanical oscillators as highly sensitive sensors. In this thesis, a novel method based on cavities in silicon substrates is introduced to detect the motion of a single motile bacterium and is compared to a highly sensitive sensor based on graphene membranes. A sensor with bacteria attached will exhibit random oscillations and these oscillations can be monitored with a laser. The rigid silicon substrates are intuitively not showing any oscillations. Nonetheless, the motility of a bacterium can still be observed by the changes in the laser path. By using an interferometric setup and monitoring the fluctuation in the voltage signal, coming from bacterial biophysical processes, the Signal-to-Noise-Ratio is determined. The amplitude increase in the fluctuation of the signal is evidently correlated with the motility of the bacteria and is demonstrated to show an increasing trend when presented with a deterministically known topographical location of the bacteria. Subsequently, the ability to detect motile cells enables the capabilities of Antibiotic Susceptibility Testing on these substrates, by monitoring the change in motion of the bacteria prior and post exposure to an effective antibiotic. The motion change, due to antibiotic exposure, is observed to show a significant difference even for a single bacterium, which opens up opportunities for an elementary non-invasive monitoring tool based on silicon.

2D flexure elements fulfil a pivotal role in the field of precision positioning systems as they do not suffer from backlash, friction or play and exhibit highly repeatable behaviour. However, planar building blocks are characterized by a limited design space as the majority of the flexure elements are flat and straight. In this thesis, it is proposed to investigate another type of flexure element that features spatial (3D) properties. More specifically, it considers the use of helical flexure geometries in motion stages. Insight into the linear kinematic behaviour of various helical surfaces with varying curvature is provided with the usage of parametric optimization, screw theory, unified stiffness method and a performance metric. The newly acquired insights served as a prerequisite for selecting a suitable topology capable of guiding a stage. For demonstration purposes, a motion stage prototype was fabricated consisting of three helical flexure elements. Additionally, an eigenfrequency analysis was performed and experimentally validated with a model vibration test. This has led to the successful realization of a helical based compliant motion stage.

Waves impacting on maritime structures can cause damage which endangers the environment around these structures. Due to high hydrodynamic loading, structures can enter the plastic regime, resulting possibly in structural failure. The understanding of the principles related to this structural failure due to waves is a key concept in this research. Due to a gap in existing literature, a simple approximation for the influence of plasticity on fluid-structure interaction is the first required step towards a better comprehension of structural failure of ship and offshore structures due to hydrodynamic loading. The choice is made to analyse a pendulum because of its ability to show clear and easy to interpret results, due to its few degrees of freedom. In an experiment, three different double pendulums are subjected to breaking waves created by dispersive focusing at three different locations in front of the pendulum. The top hinge of the pendulum is a normal hinge and the bottom hinge is a friction hinge that approximates plastic behaviour in a mechanical manner. The energy dissipated in the friction hinge, modelling plastic energy, and the energy transferred from the wave to the structure, called `total absorbed energy', decrease with increasing frictional torque. This means that when a structure allows for more plastic deformation, the energy that will be absorbed in total by the pendulum is larger. The energy transferred to the double pendulum shows more variability for the wave focused furthest away from the pendulum. A reduced-order model is constructed that can help explain the behaviour of the different pendulums.

The advent of machine learning and the availability of big data brought a novel approach for researchers to discover fundamental laws of motion. Computers allow to quickly find underlying physical laws from experimental data, without having in-depth knowledge of the system. Applications are widespread among numerous fields such as physics, chemistry, engineering, biology, climate science and finance. However, the problem is that often experimental data are polluted with noise. This causes symbolic regressors to capture the noise but not the underlying physics. In this work, we leverage the noise reduction properties of non-parametric machine learning, to improve symbolic regression methods. This combination allows for numerically robust differentiation and significantly increases the noise tolerance of common symbolic regressors. A new strategy is exemplified by combining Gaussian processes and the symbolic regression toolbox SINDy for finding governing equations from data. The method balances the quality of the fit and parsimony to avoid overfitting. The method is tested on well-known dynamical systems with varying properties. These will include the Duffing oscillator, as an example of a forced system, the Van der Pol oscillator, as an example of an autonomous system and the Lorenz attractor as an example of a chaotic system.