A Landmark Based Pinpoint Landing Simulator

Real-time position estimation for a descent lander is a critical technological need for many planned NASA missions. In particular, it enables the ability to land precisely and safely in a scientifically promising but hazardous site and is a key technology to be demonstrated by NASA in the next decade. The primary question of pinpoint landing is how to localize the lander by recognizing landmarks (craters) in the landing area and match them positively to a preexisting landmark database while the spacecraft is descending. In addition, there are several equally important operational issues. For example, what is the optimal field of view (FOV) of the descent camera to ensure maximum chance of success? What accuracy can the pinpoint landing approach achieve? How many landmarks are needed to ensure unambiguous localization? What is the performance requirement of each individual algorithm involved (the crater detection rate, the false alarm rate, ant the crater position error, speed, etc.)? In order to answer these questions, a landmark-based pinpoint-landing simulator (LAMPS) is under development. The current LAMPS can conduct two types of assessment: a position accuracy analysis and a crater constellation uniqueness analysis. The results are very valuable for pinpoint landing related R&D as well as mission design and planning. Finally, a case study of a hypothetical landing scenario is given. Introduction Craters are landforms commonly found on the surface of planets, satellites, asteroids, and other solar system bodies. A crater, in general, is a bowl shaped depression created by collision or volcanic activities. Because of their different geological ages and magnitudes of impact, craters may have a wide range of appearances. For instance, younger craters may have sharper and regular rims while aged craters may have very vague rims. Spatial densities of craters also form the primary basis for assessing the relative and absolute ages of geological units on planetary surfaces. However, a typical crater in an image has an elliptical rim and a bright to dark shading pattern, which is dictated by the lighting azimuth and elevation as well as its own topography. Because of their simple and unique geometry, craters are ideal landmarks for spacecraft navigation [2-5]. Optical landmark navigation using craters on the surface of a central body was first used operationally by the Near Earth Asteroid Rendezvous (NEAR) mission [4-5]. It has been shown to be a powerful data type for determining spacecraft orbits about the body for close flybys and low attitude orbiting. In the navigation filter, detected and identified craters were combined with Deep Space Network (DSN) radiometric tracking (Doppler and range) to estimate both orbital and asteroid physical parameters. The crater locations were also estimated. The direct benefit of optical landmarks in NEAR navigation has been enhanced navigation performance, specifically in increase in orbital position accuracy to 10 20 meter range, faster estimate convergence after maneuvers, and better solutions for dynamical parameters, such as spacecraft non-gravitational accelerations and Eros gravity perturbations. Another benefit of using landmark tracking has been the rapid determination of poorly known physical parameters of Eros which affect navigation, such as spin pole direction and spin state. Another important use for landmarks is spacecraft pinpoint landing (PPL). The current Entry, Descent and Landing (EDL) capability has a very large landing error ellipse. The landing ellipses for the Viking landers, Mars Pathfinder, Mars Polar Lander, and Mars Exploration Rovers have a semimajor axis on the order of 100-300 km. Guided entry is expected to shrink the landing ellipse to 3-6 km for second-generation landers as early as 2009. However, even if a landing ellipse is only a few kilometers in size, it is very likely to contain hazards such as craters, steep slopes, and rocks, regardless of how the ellipse is selected. To decrease the probability of landing on a hazard, one of two safe landing approaches may be employed: crater hazard detection avoidance, which will detect all hazard craters during the descent and avoid landing inside any of these craters, or pinpoint landing, which determines the lander’s position in real-time and guides the spacecraft to land at a preselected site. According to recent studies on the size/frequency of craters on the surface of Mars [7], a sufficient number of adequately sized craters for determining spacecraft position are very likely to be found in descent imagery. For example, if an image is taken at 8km above the surface with 45-degree camera Field of View (FOV), there will be an average of 94 craters of < 200 m diameter in the image. These craters can be used as landmarks to match a pre-existing crater database and, therefore, to determine the position of the lander. Figure 1: Craters in the descent image are identified and matched to a database. Using the known 3D positions of the craters and their 2D images, the position and attitude of the lander can be computed in real-time during descent. PPL Operation Scenario and Requirements The landmark-based pinpoint landing approach is as follows. First, a scientifically interesting landing site on the targeted body is selected on earth using orbital imagery, and the landmarks (e.g. craters) within the landing ellipse are mapped. During the lander descent, its initial position with respect to the landmarks as well as to the selected landing site is determined automatically on board. The lander is then guided to the landing site using continuous updates of lander position and velocity throughout the descent (Fig. 1). Three key algorithms enabling PPL are the landmark (crater) detection, landmark matching, and position estimation.