ESA GNC Conference Papers Repository
Evaluation of optical absolute navigation method using craters for lunar south pole landing
ESA desires to land on the Moon within the European Large Logistics Lander (EL3) program. EL3 comprises multiple landers, each aiming for a different landing site. To this end, the landers require a GNC system that can precisely reach any desired landing site on the entire Moon. Landing with high accuracy requires to include an absolute navigation method into the GNC system, which makes the landers MCMF pose observable. With the crater navigation system, CNav, one such method has been developed at DLR. The basic concept is first detecting impact craters in images periodically acquired of the lunar surface. The second step is matching the craters detected within that image with the ones included in an a priori determined database. From these matches the spacecrafts absolute position and orientation is determined. Our system uses three different matching strategies which are autonomously selected by considering the systems current state and other parameters. This advanced matching scheme allows for global navigation, offering a moderate measurement frequency at coarse navigation knowledge, and high-speed operation at tracking-grade state accuracy. A typical CNav operation begins with the crater detector robustly extracting craters from images of the underlying lunar surface taken by the spacecraft. Then, the crater candidates are to be matched against a crater catalog using the advanced matching scheme. The first matching technique is a form of lost-in-space matching, which in principal can be performed in the absence of any a-priori state knowledge. We call this acquisition mode. In case of better on-board navigation accuracy, e.g. from a prior successful CNav solution or from ground updates, a faster, more robust matching mode approach can be used: the tracking mode. After its successful operation, any matching method includes a thorough match verification strategy, which ensures that the probability of a false match is low. During extensive testing it was found that less than 1 percent false matches remain undetected and are returned by the method. Even then the remaining false matches can most likely be detected in a later stage by means of navigation filter internal measurement checking. DLR has a global crater database of more than 40 Mio craters available which serves as a basis for generating the on-board crater catalogs. Thus, CNav can be employed for landing everywhere on the Moon, provided sufficiently illuminated images can be taken and craters are present. Especially at the lunar south pole, it can be difficult to satisfy these two constraints. However, the south pole is one of the prime targets of future missions such as EL3. Therefore, an analysis has been performed in the context of DLRs contribution to the EL3 study to demonstrate the applicability and performance of CNav for a landing at the south pole. It has been demonstrated that viable approach trajectories exist which are sufficiently illuminated and contain sufficient craters to deliver CNav results down to altitudes of around 1 km above the landing site. In addition, a more detailed investigation of the landing conditions and their impact of the applicability of optical methods for landing on the south pole has been performed. We conclude that landing at or close to the south pole is feasible using CNav. In the paper we will present the results of both the illumination analysis and the CNav performance for a landing at the lunar south pole, thereby we will demonstrate that DLRs crater navigation can be used to land an EL3 lander at the south pole.