The current commercialisation of spaceflight significantly boosts the amount of objects in space, resulting in an increasing risk of involuntary collisions between space debris and operational spacecraft. Collisions are catastrophic for the orbital environment, as they create thousands of pieces of debris flying around that will put other satellites in danger for decades. To sustain a healthy space environment, feasible and cost-effective Active Debris Removal (ADR) and On Orbit Servicing (OOS) solutions are required that can be implemented on a wide scale. Ideal ADR and OOS solutions require no prior information on the target. This research aims to support development in that field by investigating the use of monocular-based Simultaneous Localisation AndMapping (SLAM) for pose estimation around unknown targets. The research in this work is guided by a designed technical research framework that aims to assess the robustness of monocular SLAM under realistic orbital scenarios. The pose estimator requires robustness to images with high contrast and noise, but also to adverse and changing illumination conditions. Furthermore, it requires the capability to function in dynamic environments, where objects in the background are not fixed in the target body frame. This work also analyses the hardware requirements to obtain stable performance of the pose estimator. Apart from those challenges related to space imagery, it needs to be able to obtain pose estimates at accuracies specified by the mission type. Due to the complexity and novelty of the field, spacecraft pose estimation of unknown uncooperative targets is still in early phases, as the state-of-the-art algorithms have been tested on simulations and not yet been demonstrated in real-life. In this work, The algorithm of Campos Martínez et al.[12] is adapted for spacecraft relative navigation. Background robustness is achieved by implementing a distance threshold for feature storing. Features that are located further from the camera than a predefined threshold are discarded, thereby separating fore- and background features. Additionally, pose initialisation is optimised by enhancing the robustness under low-parallax conditions, while maintaining real-time performance. Finally, the feature descriptor distance threshold is tuned to improve the accuracy of the algorithm for spacecraft relative navigation purposes. The designed pose estimator is tested on synthetically and lab-generated datasets, both capturing the Delfi-n3Xt satellite. The designed datasets allow for analysis of relative motion effects, environmental effects, like brightness changes and background objects, and for the analysis of hardware requirements. The suitability of the pose estimator for spacecraft relative navigation is assessed on three factors. Firstly, achievement of robustness to orbital relative navigation conditions is required. This includes relative motion, background, space image quality, and illumination robustness. The robustness to these factors is demonstrated in this report. The second factor is based on the capability of this system to work in real-time on space hardware. In this work, it is found that a lower-limit of 800x600 pixels on image resolution guarantees stable performance of the pose estimator. Furthermore, the frame rate should be high enough such that the target’s orientation propagation between frames does not exceed 1 deg/frame. When observing hardware used on Vision-Based Navigation (VBN) systems, it is noticed that the RNS camera in [49] could capture resolutions of 1024x1024px at a frame rate of 3 Hz. Based on the fact that the design of that system was developed more than ten years ago, and the fact that technology advances significantly over the years, it is deemed feasible that the algorithm could be implemented on space hardware. The third and last point is related to the achievable accuracy. On average, this system achieved an accuracy of sub-degree level in orientation, and centimetre-level in relative position. For particular mission phases of Rendezvous and Proximity Operations (RPO), strict accuracy requirements need to be met. For reference, the relative navigation requirements for the e.Deorbit mission [59] and the Servicing Mission 4 of the Hubble telescope [49] are observed. Since the required accuracies of both the e.Deorbit mission and Servicing Mission 4 have been achieved in this work, the designed pose estimation system is deemed suitable for spacecraft relative navigation. The novelties of this work related to robustness analysis of the implemented pose estimator to spacecraft relative motion, the capability of handling background objects in the frame, and the analysis of hardware requirements, provide a relevant contribution to the work already performed in this field.