ESA GNC Conference Papers Repository
Guidance, Navigation and Control for the autonomous rendezvous and docking of cooperative targets
The relevance of In-Orbit Servicing (IOS) missions for sustainable space exploration is growing exponentially. A wide range of applications is starting to flourish, including in-orbit refuelling and active debris removal. These missions pose tight requirements on the performance of the Guidance, Navigation, and Control (GNC) subsystem and on the autonomy level for the operation of the spacecraft. In this framework, SENER Aeroespacial is developing Rendezvous and Docking GNC based on onboard optimization technologies with the aim of increasing the flexibility, performance and reliability of IOS missions. Such techniques are already being implemented in the frame of ESA In-Space Logistics Proof-of-Concept 1 mission studies. This contribution presents a rendezvous GNC solution based on onboard convex optimization and optical navigation, which demonstrates improvements in performance, efficiency, and robustness with respect to classical approaches. Several rendezvous and capture scenarios have been assessed, focusing on cooperative and controlled targets, but also assessing non-cooperative and tumbling ones. This flexibility has been reflected in the development of an in-house generic rendezvous and capture GNC tool, the SENER Rendezvous Tool: SERVO. The guidance and control problem is formulated as a non-linear optimal control one and solved onboard by means of Sequential Convex Programming. The optimization is further exploited by implementing it in a Model Predictive Control (MPC) scheme working in real-time, improving the robustness of the control. Such implementation is based on the in-house autocodable optimization toolbox for onboard guidance, SENER Optimization Toolbox: SOTB. This optimization core has been developed ensuring independence from any toolboxes or external software elements, allowing complete visibility, verifiability and tailoring capability. The navigation architecture greatly depends on the rendezvous scenario and the degree of cooperativeness of the target. For the cooperative scenario, the relative state solution is obtained by means of the absolute state estimation of both spacecraft and is aided by optical navigation, especially for the close-range and capture phases. The chaser state knowledge is achieved by implementing classical navigation schemes, and the target state transmitted to the chaser via inter-satellite direct link is included in the navigation architecture. SERVO allows including Rendezvous Camera and LiDAR measurements in the simulation and processing them in the loop using image processing and Iterative Closest Point algorithms. These optical measurements are fused by means of a multi-rate Extended Kalman Filter, implemented in a sequential architecture. The cooperative rendezvous and docking GNC is designed, sized and simulated taking into account overall system performances and operations feasibility. The GNC is designed to ensure compatibility with the SIROM (SENER Aeroespacial docking and refuelling interface) capture requirements, in order to guarantee the feasibility of the mating scenario. Monte Carlo campaigns carried out on the SERVO System Concept Simulator demonstrate that the implemented GNC architecture is robust when subject to uncertainties in the system. Furthermore, the MPC scheme developed results in cost-efficient manoeuvres in terms of propellant consumption, when compared to classical control techniques or feedforward control architectures based on offline optimal guidance. The algorithms developed have been shown to be compatible with cooperative and non-cooperative rendezvous and docking/capture scenarios with few modifications required.