The growing space debris environment poses a significant threat to operational satellites and the future of spaceflight. To assess the collision risk between space objects, often probabilistic methods are used, which suffer from the fact that very uncertain states will lead to a low probability of collision, described as the dilution effect. Apart from this drawback, satellite operators are required to decide whether to mitigate a risk well before the time of closest approach. However, the time horizon available for this is often too short. This research aimed at studying whether novel risk assessment metrics could increase the reliability of risk quantification and extend the time horizon available for decision making. Outer probability measures have proven to be a reliable method for mitigating the dilution effect, whereas a geometry-based metric has proven more challenging to use. Together with the theoretical analysis of the metrics, their operational application has been studied.

As demand for high-bandwidth data transmission continues to increase, future satellite constellations can leverage the benefits of free space optical communication, which provides significant advantages compared to traditional radio frequency systems. Despite the development of satellite laser communication terminals (LCTs) in recent years, one major challenge remains; coupling the incoming signal into an optical fiber. In addition, most terminals use intensity modulation to encode data, which is limited in its ability to reach high data rates. So-called coherent modulation schemes offer superior performance but are not widely implemented in satellite LCT’s yet. A system that avoids fiber coupling while using coherent modulation could simplify designs, reduce complexity, and improve performance. This thesis presents the design, implementation, and evaluation of such a coherent free space optical detection system. Differential Phase Shift Keying (DPSK) was chosen as the modulation scheme due to its balance of power efficiency and implementation simplicity, delivering performance comparable to other coherent schemes with a less complex receiver architecture. The system was implemented using bulk optics and tested in a laboratory environment. A Delay Line Interferometer (DLI) was used for DPSK demodulation. The system successfully achieved a data rate of 100 [Mbit/s], demonstrating the feasibility of DPSK-based coherent free space detection. Some performance limitations were identified, with DLI instability emerging as the most notable, which is primarily caused by the long length required for this data rate. This instability resulted in intermittent bit error rate values exceeding the required threshold. These findings show that DPSK demodulation using bulk optics is favored toward higher data rates, and thus shorter delay lines. Recommendations for future work would be to operate at a higher data rate, and implement active stabilization for operational systems.

Satellites are becoming smaller and therefore their attitude sensors must reduce in size as well. However, in order to maintain 3-axis attitude determination, a combination of several sensor types is required. Moreover, the availability of these sensors is mostly dependent on the lighting conditions and slew rate of the spacecraft. In this thesis, a prototype low-budget 2-in-1 star tracker and Sun sensor was designed with similar performance but reduced technical budgets compared to a typical combination of COTS attitude sensors on CubeSats. The design can be used in both sunlight and eclipse, while stable pointing and while tumbling and it takes up less volume on a satellite. Future designs should focus on reducing solving times and power consumption by creating a dedicated microcontroller and a more optimized star tracker algorithm. Alternatively, a different CMOS chip with lower power consumption should be selected or the use of a satellite's central computing unit needs to be considered.

Greenhouse gas monitoring satellites are crucial for understanding climate change and assessing the effectiveness of climate policies. This project explores a novel concept for a metasurfaced-based spectrometer in the short-wave infrared domain. Metasurfaces are artificial planar materials with subwavelength structures that can manipulate light. This project's metasurface is designed with phase change materials to function as a reconfigurable spectral filter. Filter-based spectroscopy is realized by exploiting a compressive sensing algorithm. Using the rigorous coupled wave analysis solver of Ansys Lumerical, different metasurface designs are simulated and evaluated on their transmission response and compressive sensing performance. The final metasurfaces are obtained using particle swarm optimization with a performance metric that balances computation time and accuracy. The results show great promise for accurate and precise reconstruction of the atmospheric absorption spectrum, required for greenhouse gas monitoring.

The rapidly increasing utilization of small satellites has led to a growing demand for improved solar array technologies. Traditional solar arrays are limited by their mass and volume constraints, and thus are struggling to meet the increased demands. This feasibility study explores the potential of roll-out solar arrays as a solution to enhance the power generation capabilities of small satellites. These innovative systems are characterised by their unfurling method of deployment, and have demonstrated improved performance compared to traditional solar arrays. By analysing the feasibility of designing a roll-out solar array for small satellites, it is hoped that the operational capabilities of small satellites can be enhanced.

A comprehensive literature study revealed a significant performance gap in high power-to-mass ratios for satellites with a power range of 300 − 1000 W , leading to the development of a roll-out solar array design specifically for this power range. The study combined an analytical model with the non- dominated sorting genetic algorithm (NSGA-II), an efficient multi-objective optimisation algorithm, to generate a set of optimal designs by maximising the power-to-mass ratio and stowed power density, with the latter parameter being another key performance indicator for roll-out solar arrays. The analytical model incorporated the core design elements of a roll-out solar array like deployment mode, deployment mechanism, solar cell type, and stiffening method. A multitude of constraints ensured the feasibility of the designs, while several verification and validation tests were conducted to confirm the accuracy of the analytical model.

The optimisation results revealed that a variety of design combinations exist that could outperform the state of the art solar array capabilities. Optimal designs demonstrated power-to-mass ratios exceeding 145 W /kg and stowed power densities over 55 kW /m3. These designs would deliver over double the power-to-mass ratio of traditional solar arrays in the 300 − 1000 W range, while remaining comparable to the 120 − 140 W /kg state of the art performance of arrays outside of this range. Furthermore, optimal designs offer up to 40% increases in stowed power density compared to the 40 kW /m3 state of the art benchmark. Despite the promising gains, the study reveals some key limitations to roll-out solar arrays. These are the increased development costs, the higher risks, and their bulky geometry being potentially inefficient for smallsats. Ultimately, the study provides for a strong foundation for the development of a roll-out solar array. The results confirm the potential of such a system, and future research is recommended to solidify the feasibility of this innovative technology.

Future Mars exploration missions require safe and controlled landing on the planet’s surface. Conventional entry, descent and landing (EDL) technologies, such as parachutes and rigid aeroshells, face limitations in meeting increasing demands for heavier payloads, harsher entry conditions, and desired landing locations due to their limited deployment windows or geometry constraints set by current launchers. The stacked-toroid inflatable aerodynamic decelerator (IAD) has emerged as a promising EDL technology which has the potential to enable new and ambitious applications. Unlike conventional aeroshells, it utilizes flexible, high-temperature resistant materials that can be folded during orbital injection and transportation, and subsequently deployed before entering the Martian atmosphere. To address the complex interdependence of design variables and the multidisciplinary nature of stacked-toroid analysis, this research proposes a novel Multidisciplinary Design Analysis and Optimization (MDAO) framework. The framework integrates aerodynamics, aerothermodynamics, structural analysis, mass estimation, and Flexible-Thermal Protection System (F-TPS) sizing with trajectory simulations for a parametrized stacked-toroid. The major design variables are parameterized to trace model responses back to the design space. An additional novel contribution is the inclusion of a smaller torus on the IAD’s shoulder. For aerodynamics and aerothermodynamics modelling, the local-inclination panel method is implemented, separately addressing the continuum, rarefied, and transitional flow regimes using well-established analytical methods and bridging functions. The scalloping phenomenon of the deflected surface is accounted for through semi-empirical expressions and experimentally-fitted correlations, capturing the additional aerodynamic and aerothermal contribution. F-TPS sizing employs a 1D Finite Difference Method (FDM) with harmonic weighted averaging of material characteristics to accommodate abrupt thickness variations. Results from each discipline are compared to experimental, flight, and high-fidelity numerical data, showing consistent agreement under various conditions. All disciplines present mean percentage errors in the order of 10-20% which are deemed acceptable for early design stages. The framework is applied to the ESA MiniPINS study, to demonstrate its applicability to a novel EDL architecture for a penetrating probe. The proposed environment efficiently evaluates the stacked-toroid optimum design for minimum mass, weighing only 3.72 kg whilst complying with mission requirements and optimisation constraints. The design remains robust against variations in entry parameters and atmospheric density, requiring minor adjustments for the penetrator impact speed. The framework enables design space exploration, revealing trends favoring stacked-toroids with low inner torus radii and large numbers of tori to minimize aerothermal loads while ensuring sufficient aerobraking. The inflated radius remarkably results in a global variable that can further simplify the design space to a single input for near-optimum rapid evaluations. The feasible parameter ranges identified through the design space search expedite the evaluation and optimization of stacked-toroids’ multidisciplinary performance, aiding decision-making in early design stages of future Mars missions.

The accurate prediction of satellite orbits is essential for the proper functioning of space-based services such as navigation, communication, and Earth observation. However, atmospheric drag is a significant source of error in satellite orbit prediction, especially in Low Earth Orbit (LEO), where the majority of space objects operate. The thermosphere, the outermost layer of Earth's atmosphere, plays a crucial role in determining atmospheric drag. However, the thermospheric density is subject to high levels of uncertainty, up to 30%, when computed from atmospheric models. This uncertainty is particularly relevant in LEO, where it can affect the operational lifespan of satellites.

To increase the accuracy of thermospheric density models, this paper presents an assessment of the accuracy of orbit predictions using empirical Thermospheric Mass Density (TMD) observations obtained from the Swarm C satellite. The study uses a Principal Component Analysis (PCA) to decompose a fine grid of density into the main temporal and spatial modes. Then, each of the modes is calibrated with Swarm C observations using a Least Squares Estimation (LSE) algorithm. The calibration is validated with the observation using unbiased metrics. These observations were assimilated into the NRLMSISE-00 atmospheric model, and the resulting calibrated density model was used to predict the orbits of Swarm C, GRACE-FO and Sentinel 1-A satellites. To analyse the consistency of the results, a slicing window analysis was performed, and the median evolution of the windows was computed.

The results of the study indicate that a calibrated density model with a several satellite geometries, such as the cannonball model, a panel model, or a scaled panel model, can significantly improve the accuracy of orbit predictions in LEO. During March 2022, a period of medium solar activity, the median accuracy of orbit predictions for Swarm C was reduced from 20.67 km with NRLMSISE-00 to 13.75 km with the calibrated model and a scaled panel model geometry. During the same period, the median accuracy of orbit predictions for GRACE-FO C was reduced from 17.98 km with NRLMSISE-00 to 2.89 km with the calibrated model and an (unscaled) panel model geometry. These findings have important implications for the sustainable operation of satellites in the increasingly crowded space environment.

The Kuiper belt is one of the last mostly unexplored regions in the Solar System. Exploration of the Kuiper belt can greatly increase humanity's understanding of the Solar System's formation and evolution. The use of low-thrust propulsion for Kuiper belt missions has the potential to improve the payload mass of the mission due to the high efficiency of its propellant. This research looks at the required methodology to optimize low-thrust trajectories with Kuiper belt object flybys. The trajectory is modelled using Tudat with spherical shaping as the trajectory parameterization method. A methodology is constructed which switches between high-thrust and low-thrust legs to constrain the input space for the low-thrust optimization problem. By using close-approach graphs and optimizing multiple flybys at the same time a trajectory with two Kuiper belt object flybys is found. The result is a robust method to find Kuiper belt object flyby trajectories with low-thrust propulsion.

Access to microgravity has benefits across a large variety of fields. It presents unique features including the absence of convection, sedimentation, and buoyancy, which introduce new possibilities within physical and chemical processes. The dominance of diffusion in microgravity allows more uniform and precise structures to be formed within material sciences. Moreover, the increased influence of surface tension enhances the precision of adhesion, contact, and material interactions, resulting in the production of intricate and higher-quality products. The access to true microgravity is extremely limited, however, we can utilise platforms for simulated microgravity on ground, such as the Random Positioning Machine (RPM). This provides a cost-effective and accessible alternative for gravitational research. To describe the RPM's limitations and performance, a complete kinematic model was developed. This allows new guidelines for its use and extends the access to microgravity across disciplines.

To be able to understand and predict space weather better, global in-situ satellite measurements are needed at an altitude of 100-150 km. However, due to the high density, space weather induced winds and the lack of data, it is a challenge to design a satellite mission to that region. Therefore, in this thesis, the influence of horizontal wind on satellite aerodynamics in the lower thermosphere is analyzed. The impact of the choice in satellite geometry design and orbital parameters on the satellite's aerodynamics and wind sensitivity is tested. To model satellite aerodynamics in this high-density region, a method of using the Stochastic PArallel Rarefied-gas Time-accurate Analyzer, SPARTA, is proposed. It is proven that the worst-case horizontal wind can have a negligible influence on the satellite’s drag coefficient depending on its attitude and satellite geometry. Based on the obtained conclusions, recommendations are given to simplify the design process and reduce aerodynamic drag.

In recent decades, the field of autonomous driving has witnessed rapid development, benefiting from the development of artificial intelligence-related technologies such as machine learning. Autonomous perception in driving is a key challenge, in which multi-sensor fusion is a common feature. Due to the high resolution and rich information, the camera is one of the core perceptual sensor in autonomous systems. However, the camera provides no knowledge on distance (or depth), which is insufficient for the requirements of autonomous driving. On the other hand, LiDAR provides accurate distance measurements, however the information is sparse. The complementary characteristics of cameras and LiDAR have been exploited over the past decade for autonomous navigation. In order to be able to fuse the camera and LiDAR sensor system jointly, an efficient and accurate calibration process between sensors is essential. Conventional methods for calibrating the camera and LIDAR rely on deploying artificial objects, e.g., checkerboard, on the field. Given the impracticality of such solutions, targetless calibration solutions have been proposed over the past years, which require no human intervention and are readily applicable for various autonomous systems, e.g., automotive, drones, rovers, and robots.

In this thesis, we review and analyze several classic targetless calibration schemes. Based on some of their shortcomings, a new multi-feature workflow called MulFEA (Multi-Feature Edge Alignment) is proposed. MulFEA uses the cylindrical projection method to transform the 3D-2D calibration problem into a 2D-2D calibration problem and exploits a variety of LiDAR feature information to supplement the scarce LiDAR point cloud boundaries to achieve higher features similarity compared to camera images. In addition, a feature matching function with a precision factor is designed to improve the smoothness of the objective function solution space and reduce local optima. Our results are validated using the open-source KITTI dataset, and we compare our results with several existing targetless calibration methods. In many different types of roadway environments, our algorithm provides more reliable results regarding the shape of the objective function in the 6-DOF space, which is more conducive for the optimization algorithms to solve. In the end, we also analyze the shortcomings of our proposed solutions and put forward a prospect for future research in the field of joint camera-Lidar calibration algorithms.

In this thesis, a new method to approximate the cost function of Low-Thrust, Multiple-Gravity-Assist interplanetary trajectories using a Machine Learning surrogate is proposed. This method speeds up the optimization process without fine tuning of the surrogate parameters for every individual case. The computational cost of obtaining training data was identified as the main limitation when using Machine Learning methods for this purpose. Therefore, the surrogate was built with an Online Sequential Extreme Learning Machine Multi-Agent System (OS-ELM-MAS) due to its theoretical good performance when the training data is limited. A high-fidelity global optimization problem was implemented, and a method to include the surrogate during the optimization process was designed. This method does not require specialized optimization algorithms. The parameters that control the interaction between the surrogate and the optimization process were identified and a procedure to obtain the best values was designed and applied. The final results show that the use of the surrogate improves the optimization results when evaluations of the cost function are computationally expensive. However, the values of the parameters that control the interaction between the surrogate and the optimization algorithm had to be carefully selected. The search for a general procedure to obtain these parameters without repeated tests is proposed for future research. Several applications to new optimization problems of the method developed in the thesis are also proposed for future research.

This thesis work has been motivated by the growing interest in small-spacecraft small-body missions and more specifically inspired by a proposed mission to the largest Martian Trojan (5261) Eureka. This binary asteroid represents a promising target for a planetary mission, as a better understanding of its formation and evolution would provide valuable insights into the history of both the Martian system and binary asteroids. This work has focused on identifying orbital designs which would optimise the geodetic parameters estimate, and thus best help characterise the asteroid's internal structure. The orbit determination solutions have been computed from simulated tracking data in various orbital configurations. To compensate the highly perturbed nature of the binary asteroid's environment, pseudo-periodic solutions identified around Eureka have been used as stable initial conditions for the spacecraft orbits. As Eureka's shape, gravity field and rotational state are not well characterised yet, large resulting uncertainties have been accounted for. They have been propagated to assess the robustness of the spacecraft orbits and subsequent orbit determination solutions to dynamical mismodelling, in order to eventually limit the science risk of the mission. For single CubeSat configurations, an optimal semi-major axis has been identified, being the one closest to the asteroid before orbital instability would start deteriorating the orbit determination solution. Non-equatorial retrograde orbits have been found to provide the lowest formal errors for geodetic parameters. Moreover, the benefit of simultaneously deploying two or even three CubeSats around the asteroid has been demonstrated. In both the two and three CubeSats configurations, promising orbital designs have been identified. They provide an estimation of the geodetic parameters which is accurate enough to bring insights into Eureka’s internal structure, and has been proven to be robust to dynamical mismodelling.
The methodology developed in this work is of interest to design other small-spacecraft small-body missions, and optimise their achievable geodetic parameters estimate. Additionally, Eureka's dynamical model and the large uncertainties assigned to it have been mostly based on models available for other binary asteroids. This brings confidence in the possible applicability of the identified optimal orbital designs to CubeSats missions targeting other binary asteroids.

Large Low Shear-Velocity Provinces (LLSVPs) in Earth have been recognized in seismic data for several decades. The anomalies are however still poorly understood for reasons including their inconvenient location in the deep mantle, inhibiting progress in our understanding. Comprehension of these structures is important, because of their speculated major role in the formation and evolution of the Earth, and her residents. The gravity anomaly associated with the LLSVPs, the C22-lobs, are roughly observed in the gravity field of Mercury too. Therefore, the gravity and topography data about Mercury is assessed for indications of similar deep mantle features as observed in Earth. The advantage of Mercury is its thin mantle, that allows for a relatively shallow location of the core-mantle boundary (CMB), hence shallow deep mantle anomalies. The cause of the thin mantle is yet unknown, currently suggested explanations include vaporization of the mantle due to the immense heat of the solar nebula or a hit-and-run collision scenario that ripped off the mantle of Mercury. In this research it is investigated how deep mantle structures like LLSVPs and CMB topography could account for the gravity field of Mercury. It is concluded that deep mantle structures with comparable size and density characteristics as the LLSVPs in Earth could exist. Compensation by CMB topography would require a rough CMB. The results are highly affected by the limited knowledge about crustal compensation in Mercury and the truncation of the gravity field. Better crustal models of Mercury could be obtained by investigating local gravity anomalies in relation to topographic features, considering flexural isostasy and using normal modes. In addition to that, BepiColombo is currently en route to Mercury to gather more global data of Mercury, that would highly aid in our understanding of the crust and mantle of Mercury. A relation comparable to the hypothesised relation between hotspot volcanism and LLSVPs on Earth is also investigated for Mercury. A relation was not necessarily observed, however it was also not ruled out and requires more attention as soon as the crustal models have improved and more global coverage of volcanic features on Mercury is available.

In interplanetary mission design, ballistic capture is the phenomenon by which a spacecraft approaches its target body, and performs a number of revolutions around it, without requiring manoeuvres in between. For a spacecraft to be captured, its gravitational interaction with at least two celestial bodies has to be taken into account. Because of their fail-safe nature (eliminating the possibility of single point failures), their fuel efficiency and their wider launch windows, ballistic capture trajectories are of particular scientific and engineering interest. Capture orbits are characterized by a specific qualitative dynamics, defining almost-invariant regions in a given space, guiding transport phenomena; the introduction of structures, defining and bounding such domains, naturally follows. Traditionally, the set of initial conditions leading to capture, the Capture Set, has been computed by sampling the domain of interest, and hence analysing the forward and backward behaviour of the orbit associated to each sample. The main limitations of this approach reside in its large computational cost and, even for a dense grid, in the non-smooth approximation of the aforementioned boundary regions; the theory of Lagrangian Coherent Structures (LCS) has the potential of overcoming both limitations, allowing at the same time for a more insightful description of the phenomenon. In fact, Lagrangian Coherent Structures identify transport barriers in dynamical systems, separating regions with qualitatively different dynamics. The development of heuristics applicable to ballistic capture trajectory design and informed by such theory (i.e. flow-informed) appears desirable. In this research, different flow-informed approaches are presented and their relations with ballistic capture are discussed: following, a new heuristic, the Stroboscopic Strainline, is introduced. This new tool is therefore applied to different case studies at Mars, in order to approximate the capture sets associated to different numbers of revolutions and geometries. While a real-ephemerides model has been used to model the dynamical environments, different levels of fidelity have been investigated: perturbing forces have been introduced not only to obtain more accurate results, but also to test the robustness of the proposed technique with respect to different features of the underlying dynamical model. Finally, it is shown how Stroboscopic Strainlines are a good candidate for characterizing the qualitative behaviour of ballistic trajectories, both forward and backward in time.

Building on recent advances in the fields of low-thrust trajectory optimization based on shaping methods, Artificial Neural Networks, and surrogate models in Evolutionary Algorithms, an investigation into a novel optimization routine is conducted. A flexible Python tool to evaluate linked trajectories in a two-body model based on the hodographic shaping method is implemented and used to develop an evolutionary optimization approach where a Genetic Algorithm is assisted in finding new candidate solutions by a surrogate model. This surrogate is constructed from previous fitness function evaluations using Machine Learning, specifically by training an Artificial Neural Network. After deriving suitable (hyper-)parameters for the Genetic Algorithm and the Artificial Neural Network an experimental investigation into the algorithm's performance is conducted with a focus on the design of the surrogate for low-thrust trajectory problems. Two example problems based on the Dawn trajectory and the GTOC2 problem are studied. The surrogate approach is able to find good new candidate solutions, i.e. solutions that improve the population's overall fitness, especially when the surrogate is designed to approximate the shaping computation. Additionally, the use of a surrogate pretrained on a general data set of low-thrust transfers is tested and found to considerably improve the initial quality of the model, meaning that more good candidate solutions are found early on, accelerating the algorithm's convergence.

Research on laser communication technology for satellites has led to a sufficient understanding of this technology to start working on commercial applications. In this work a design is proposed for a lasercom system for CubeSats, a type of nanosatellites that is used increasingly often for commercial missions. The goal of this design is to offer substantially higher datarates from low earth orbit to ground compared to radio links, without regulatory constraints and at competitive costs. To support the first-order design as proposed in this work, a detailed implementation is described. One component of this design, the sensor that accurately measures the angle of the incoming light, is analysed experimentally.