Multi-View ICP for 3D reconstruction of unknown space debris

With a growing threat from space debris, it has become more important for space agencies to have a means of reducing the probability of escalating the dangers posed by it. For this purpose, servicer satellites need to be built. These satellites need to be able to interfere with “dead” satellites and deorbit defunct satellites. In order to be able to navigate around these types of objects, it is necessary for the servicer satellite to be able to determine its position around the object and the relative motion between the debris and the servicer. This thesis describes an algorithm that determines the trajectory of the servicer satellite based on a 3D reconstruction of a model of the target object. It takes 3D point clouds from a sensor and transform these point clouds into a 3D model that can be used by the servicer to navigate and determine its position around the target object. The approach uses a simulation of a LIDAR instrument to obtain the data which is then used as input for the algorithm to reconstruct the path of the servicer satellite. This thesis was done at Airbus Defense and Space as it is currently activley trying to develop the capability to interfere with space objects. The resulting algorithm can perform a 3D reconstruction of a target object in real-time if appropriate parameters are being used and it can also determine the position of the servicer around this model. Moreover, it can also reconstruct the trajectory for the observer around its target.
Engineering
Space
Aerospace Engineering