Marine pollution is a critical issue impacting the global community, with underwater waste a particularly daunting challenge. While autonomous detection and collection of underwater waste is highly desirable, these are extremely difficult tasks. This difficulty arises from the intrinsic complexities of the aquatic environment, including variable lighting conditions, reduced visibility, and the complex nature of water currents. This paper focuses on novel approaches for autonomous detection of underwater waste and proposes to incorporate domain knowledge to refine deep-learning-based underwater object detection techniques. More specifically, the domain knowledge is represented with models in the state space form that describe the motion of the target objects to assist the classification of objects. Moreover, optical flow is combined with a k-means clustering algorithm to extract the trajectory of the target objects from videos. These trajectories are subsequently fed into a neural network that is trained using knowledge distillation, enhanced with domain knowledge. For our experiments, a simulator is devised to facilitate the creation of a dataset for developing and testing the proposed architecture. The results of the experiments demonstrate that including domain knowledge within the object detection approach with neural networks provides numerous substantial advantages, including enhanced robustness against noisy and poorly labelled data, facilitation of semi-supervised learning, and consistent superiority in accuracy over the baseline scenario. Additionally, combining the domain knowledge with a neural network significantly increases the computational speed of object detection compared to using the standalone domain knowledge module.

Assembly, Integration and Verification (AIV) in space makes launching geosynchronous satellites faster and significantly cheaper in the long term. A space-tug is launched into space to perform AIV there. It assembles a standardised satellite consisting of several modules. The modules are designed in such a way that the required subsystems for a communication satellite are incorporated in the modules. Examples of these modules are a propulsion module, a solar array module and a computer module. Due to the standardised modules, testing time and costs can be reduced significantly. This ensures a delivery time of maximum one year, which is the time from order until operations in space. The modules are efficiently packed and connected to external beams in the launch vehicle, to make sure that two satellites can be launched simultaneously. The external beams take up the extreme loads that occur during launch. This decreases the dry mass of the satellite, as the modules do not need as much structural mass. The subsystem design and structural analysis result in a drymass of 1847 kg per satellite. Next to the two satellites, a refuelling tank is added in the launch vehicle to refuel the tug. The tug requires 2921 kg of fuel to transfer the two satellites and go back to its initial state. Due to the modularity of the satellites, the lifetime of the satellites can be increased. Regarding the economic feasibility of the mission, a full return on investment is expected after 15 years of operations in base case scenario.