DLR demonstrates rendezvous using GPS and Optical Navigation
Remarkable about this experiment is the fact that no GPS data of Tango was used during the complete experiment's duration. The relative navigation was based on angles-only measurements extracted from images collected by Mango. The successful conclusion of ARGON shows that guidance, navigation and control techniques based on optical navigation available at DLR/GSOC are qualified and ready to support on-orbit servicing rendezvous scenarios of strategic relevance.
Finally the Advanced Rendezvous demonstration using GPS and Optical Navigation (ARGON) have been conducted successfully in the time frame April 23-27, 2012. Since August 2011 the control of PRISMA is back to OHB-SE in Solna, Sweden, thus the mission operations and experiment supervision was done from the Swedish mission control center. The experiment team itself consisted of 5 members of the Space Flight Technology department of DLR/GSOC.
The goal of ARGON was to demonstrate a rendezvous to a non-cooperative passive target, which played the role of a client s/c in a possible on-orbit servicing scenario. As a consequence, the experiment team acted such as no information was available about Tango, i.e., no knowledge about its orbit and no usage of GPS data. In an attempt to mimic a realistic scenario, even a rotation was introduced on Tango to emulate an uncontrolled or out-of-order object. On the contrary, Mango played the role of the servicer s/c in an on-orbit servicing scenario. Its GPS data was processed by the well-established Precise Orbit Determination (POD). But how did we know where Tango was? This is the innovation here: the formation parameters describing the relative position between Mango and Tango were computed based on images and thereof extracted angles-only measurements. For this purpose it was very important to keep Tango in the field-of-view of Mango’s VBS camera system (star-tracker). This was accomplished through a chain of tasks which were executed in a ground-in-the-loop fashion by the experiment team:
- the image processing,
- the angles-only relative navigation,
- the formation guidance strategy, and
- the corresponding maneuver planning.
At the beginning of ARGON the initial formation between Mango and Tango was not known to the experiment team. Based on Two-Line Elements (TLE) of Tango and GPS data of Mango coarse formation parameters were computed and an along-track distance of roughly 30 km was first guessed. In parallel, the VBS system was switched on in order to collect continuous images of Tango. A first small maneuver was planned and commanded to (a) start a drift towards the target satellite and (b) improve the results of the following relative navigation solution. During the subsequent ground contact the pictures collected on-board were downloaded and processed on-ground. Based on the extracted angles-only measurements the relative navigation was performed for the first time. This updated knowledge about the formation was taken into account for the next maneuver planning and so on. Following this strategy, maneuvers were planned twice a day and executed to keep the formation close to the guidance plan. The ultimate aim was to bring Mango to a hold-point in 3 km along-track distance from Tango in a safe relative orbit. Figure 1 shows an overlap of example images over one orbit during ARGON. The time between 2 images is circa 10 minutes in that example and one can clearly identify the almost circular relative orbit of Mango with respect to Tango.
Fig 1: Overlap of example images of Tango over one orbit taken from Mango every 10 minutes
For safety reasons the Swedish colleagues also used Tango's GPS data for the formation monitoring to guarantee a secure approach, but never communicated any of this information to the GSOC staff. Only in the presence of a high collision risk, they were allowed to interfere and terminate the experiment to keep the satellites alive, but this was never necessary.
After the completion of ARGON the dumped GPS data of Tango was finally processed and the relative geometry during the complete experiment time was available to the experiment team as well. The actual development of the formation during ARGON is depicted in Figure 2. The comparison to the optical relative navigation showed extraordinary accuracies with errors below 5-10 m over the complete rendezvous on all relative orbital elements (excluding the along-track distance).
Fig. 2: Actual relative position of Mango mapped on the orbital frame centered on Tango (origin) computed from GPS POD
For further information about the experiment in general, the developed flight dynamics system and the achieved results refer to [1] and [2]. The gained experience is considered as very valuable for future missions like DEOS, where on-orbit servicing shall be further developed and realized. The demonstrated guidance, navigation and control techniques based on optical navigation available at DLR/GSOC are qualified and ready to support on-orbit servicing rendezvous scenarios of strategic relevance.
[1] D'Amico, S. et al.: "Non-Cooperative Rendezvous using Angles-only Optical Navigation: System Design and Flight Results", Journal of Guidance, Control and Dynamics (JGCD), 2012.
[2] D'Amico, S. et al.: "Flight Demonstration of Non-Cooperative Rendezvous using Optical Navigation", ISSFD Conference Paper, 2012.
Tags: