Vision-based navigation solutions require training and validation datasets that are as close as possible to real images. Our team and partners develop computer vision algorithms for space exploration (Mars, Jupiter, asteroids, the Moon), and for in-orbit operations (rendezvous, robotic arms, space debris removal). There is a new wave of missions targeting cislunar orbit or the Moon surface. Of course "real images" are rarely available before the mission. Ground-based test facilities such as robotic test benches embarking mock-ups or experiences with scaled mission analogues (mars terrain analogue, drones flights, etc.) are useful, yet they are limited.

Recent interest in on-orbit servicing and Active Debris Removal (ADR) missions have driven the need for technologies to enable non-cooperative rendezvous manoeuvres. Such manoeuvres put heavy burden on the perception capabilities of a chaser spacecraft. This paper demonstrates Convolutional Neural Networks (CNNs) capable of providing an initial coarse pose estimation of a target from a passive thermal infrared camera feed. Thermal cameras offer a promising alternative to visible cameras, which struggle in low light conditions and are susceptible to overexposure. Often, thermal information on the target is not available a priori; this paper therefore proposes using visible images to train networks. The robustness of the models is demonstrated on two different targets, first on synthetic data, and then in a laboratory environment for a realistic scenario that might be faced during an ADR mission. Given that there is much concern over the use of CNN in critical applications due to their black box nature, we use innovative techniques to explain what is important to our network and fault conditions.