Precise orbit determination through a joint analysis of optical and radiometric data

The next-generation space robotic missions will require autonomous probes capable to carry out complex navigation tasks with limited human input. Innovative data-fusion techniques are currently under development to support real-time spacecraft navigation operations through a combined processing of multi-sensor datasets, including, for example, radio tracking and optical measurements. Image-based data can provide crucial information to improve the localization of orbiters through the detection of features displaced across planetary surfaces and their registration with an onboard database. In case of lunar missions, craters represent a key dataset to improve the reconstruction of the spacecraft trajectory. In addition to describing the machine vision techniques used to detect and identify craters in the onboard images, we present here numerical simulations based on a lunar orbiter to investigate the strengths of the multi-sensor orbit determination approach.