AI-assisted hazard detection for safe lunar landing

The return of humans to the Moon will require increased capability beyond that of the previous Apollo missions. Longer stay times and a greater flexibility with regard to landing locations are among the many improvements planned. A descent and landing system that can land the vehicle more accurately than Apollo with a greater ability to detect and avoid hazards is essential to the development of a Lunar outpost, and also for increasing the number of potentially accessible Lunar sortie locations. This descent and landing system should allow landings in more challenging terrain and provide more flexibility with regard to mission timing and lighting considerations, while maintaining safety as the top priority. The lunar landing system under development by the ALHAT (autonomous landing and hazard avoidance technology) project is addressing this by providing terrain-relative navigation measurements to enhance global-scale precision, an onboard hazard detection system to select safe landing locations, and an autonomous GNC (guidance, navigation, and control) capability to process these measurements and safely direct the vehicle to a landing location. This landing system will enable safe and precise lunar landings without requiring lunar infrastructure in the form of navigation aids or a priori identified hazard-free landing locations. The safe landing capability provided by ALHAT uses onboard active sensing to detect hazards that are large enough to be a danger to the vehicle but too small to be detected from orbit a priori. Algorithms to interpret raw active sensor terrain data and generate hazard maps as well as identify safe sites and recalculate new trajectories to those sites are included as part of the ALHAT System. These improvements to descent and landing will help contribute to repeated safe and precise landings for a wide variety of terrain on the Moon.

In the area of planetary landing, hazard detection and avoidance is the act of driving a vehicle to a safe landing area using onboard resources. A hazard detection sensor is used to scan the terrain and these measurements are evaluated to determine where the safe landing sites are located. The selected site is generally not the same as the nominal target, so the vehicle must divert to the new site. This activity involves the interaction between several components, including a suite of onboard GNC algorithms that work together to efficiently choose and divert to a new site. This paper presents the Hazard Boresight Relative Navigation concept, which is a method that provides a common interface between the hazard scan, safe-site selection algorithm, size of the target-relative landing ellipse, divert offset distance and guidance targeting algorithm. After the safe-site is selected from the hazard scan, the original inertial target is replaced with a vehicle-relative target, which is initialized by a measurement from the hazard scan. The new target-relative position state is estimated over time in the navigation filter, and is fed to the guidance algorithm to perform the divert maneuver. In addition to detailing the Hazard Boresight Relative Navigation concept, this paper also presents some general landing terms that can be used in the greater discussion, as well as analysis on how to estimate and predict the vehicle footprint dispersion ellipse during flight, which is used in the safe-site selection algorithm.

A second generation, compact, real-time, air-cooled 3-D imaging Flash Lidar sensor system, developed from a number of cutting-edge components from industry and NASA, is lab characterized and helicopter flight tested under the Autonomous Precision Landing and Hazard Detection and Avoidance Technology (ALHAT) project. The ALHAT project is seeking to develop a guidance, navigation, and control (GN&C) and sensing system based on lidar technology capable of enabling safe, precise crewed or robotic landings in challenging terrain on planetary bodies under any ambient lighting conditions. The Flash Lidar incorporates a 3-D imaging video camera based on Indium-Gallium-Arsenide Avalanche Photo Diode and novel micro-electronic technology for a 128 x 128 pixel array operating at a video rate of 20 Hz, a high pulse-energy 1.06 μm Neodymium-doped: Yttrium Aluminum Garnet (Nd:YAG) laser, a remote laser safety termination system, high performance transmitter and receiver optics with one and five degrees field-of-view (FOV), enhanced onboard thermal control, as well as a compact and self-contained suite of support electronics housed in a single box and built around a PC-104 architecture to enable autonomous operations. The Flash Lidar was developed and then characterized at two NASA-Langley Research Center (LaRC) outdoor laser test range facilities both statically and dynamically, integrated with other ALHAT GN&C subsystems from partner organizations, and installed onto a Bell UH-1H Iroquois Huey helicopter at LaRC. The integrated system was flight tested at the NASA-Kennedy Space Center (KSC) on simulated lunar approach to a custom hazard field consisting of rocks, craters, hazardous slopes, and safe-sites near the Shuttle Landing Facility runway starting at slant ranges of 750 m. In order to evaluate different methods of achieving hazard detection, the lidar, in conjunction with the ALHAT hazard detection and GN&C system, operates in both a narrow 1deg FOV raster-scanning mode in which successive, gimbaled images of the hazard field are mosaicked together as well as in a wider, 4.85deg FOV staring mode in which digital magnification, via a novel 3-D superresolution technique, is used to effectively achieve the same spatial precision attained with the more narrow FOV optics. The lidar generates calibrated and corrected 3-D range images of the hazard field in real-time and passes them to the ALHAT Hazard Detection System (HDS) which stitches the images together to generate on-the-fly Digital Elevation Maps (DEM's) and identifies hazards and safe-landing sites which the ALHAT GN&C system can then use to guide the host vehicle to a safe landing on the selected site. Results indicate that, for the KSC hazard field, the lidar operational range extends from 100m to 1.35 km for a 30 degree line-of-sight angle and a range precision as low as 8 cm which permits hazards as small as 25 cm to be identified. Based on the Flash Lidar images, the HDS correctly found and reported safe sites in near-real-time during several of the flights. A follow-on field test, planned for 2013, seeks to complete the closing of the GN&C loop for fully-autonomous operations on-board the Morpheus robotic, rocket-powered, free-flyer test bed in which the ALHAT system would scan the KSC hazard field (which was vetted during the present testing) and command the vehicle to landing on one of the selected safe sites.

Safe planetary landing is considered a key technology for future robotic and manned planetary landing missions. The relay hazard detection and proportion–integration–differentiation avoidance guidance algorithms were used in Chang’e-3 mission, which not only increased the complexity of the guidance system, but also resulted in non-fuel-optimal avoidance guidance from the viewpoint of fuel consumption. To further develop and improve the hazard detection and avoidance scheme of Chang’e-3, novel autonomous hazard avoidance methodologies should be investigated. This paper addresses an innovative hazard detection and avoidance scheme for safe lunar landing from the following three aspects: imaging flash lidar based hazard detection, safe landing site selection strategy, and minimum-fuel hazard avoidance guidance. First, the three-dimensional imaging flash lidar, a developing three-dimensional imaging sensor, is utilized to rapidly and precisely detect three-dimensional terrain of the landing area. Second, the hazard detection and optimum landing site selection strategy inherited from Chang’e-3 are improved and enhanced to estimate the potential obstacles, and select an optimum landing site which is the guidance target of following hazard avoidance. Next, the fuel-optimal hazard avoidance guidance problem is transcribed into as a minimum-fuel consumption optimization problem using the Gauss pseudospectral method, which is easily solved by the open-source software GPOPS. Finally, the validity of the autonomous hazard detection and avoidance guidance scheme proposed in this paper is confirmed by computer simulation.

The Hazard Detection System (HDS) is a component of the ALHAT (Autonomous Landing and Hazard Avoidance Technology) sensor suite, which together provide a lander Guidance, Navigation and Control (GN&C) system with the relevant measurements necessary to enable safe precision landing under any lighting conditions. The HDS consists of a stand-alone compute element (CE), an Inertial Measurement Unit (IMU), and a gimbaled flash LIDAR sensor that are used, in real-time, to generate a Digital Elevation Map (DEM) of the landing terrain, detect candidate safe landing sites for the vehicle through Hazard Detection (HD), and generate hazard-relative navigation (HRN) measurements used for safe precision landing. Following an extensive ground and helicopter test campaign, ALHAT was integrated onto the Morpheus rocket-powered terrestrial test vehicle in March 2014. Morpheus and ALHAT then performed five successful free flights at the simulated lunar hazard field constructed at the Shuttle Landing Facility (SLF) at Kennedy Space Center, for the first time testing the full system on a lunar-like approach geometry in a relevant dynamic environment. During these flights, the HDS successfully generated DEMs, correctly identified safe landing sites and provided HRN measurements to the vehicle, marking the first autonomous landing of a NASA rocket-powered vehicle in hazardous terrain. This paper provides a brief overview of the HDS architecture and describes its in-flight performance.

For the first time, a suite of three lidar sensors have been used in flight to scan a lunar-like hazard field, identify a safe landing site, and, in concert with an experimental Guidance, Navigation, and Control (GN&C) system, guide the Morpheus autonomous, rocket-propelled, free-flying test bed to a safe landing on the hazard field. The lidar sensors and GN&C system are part of the Autonomous Precision Landing and Hazard Detection and Avoidance Technology (ALHAT) project which has been seeking to develop a system capable of enabling safe, precise crewed or robotic landings in challenging terrain on planetary bodies under any ambient lighting conditions. The 3-D imaging flash lidar is a second generation, compact, real-time, air-cooled instrument developed from a number of cutting-edge components from industry and NASA and is used as part of the ALHAT Hazard Detection System (HDS) to scan the hazard field and build a 3-D Digital Elevation Map (DEM) in near-real time for identifying safe sites. The flash lidar is capable of identifying a 30 cm hazard from a slant range of 1 km with its 8 cm range precision at 1 sigma. The flash lidar is also used in Hazard Relative Navigation (HRN) to provide position updates down to a 250m slant range to the ALHAT navigation filter as it guides Morpheus to the safe site. The Doppler Lidar system has been developed within NASA to provide velocity measurements with an accuracy of 0.2 cm/sec and range measurements with an accuracy of 17 cm both from a maximum range of 2,200 m to a minimum range of several meters above the ground. The Doppler Lidar's measurements are fed into the ALHAT navigation filter to provide lander guidance to the safe site. The Laser Altimeter, also developed within NASA, provides range measurements with an accuracy of 5 cm from a maximum operational range of 30 km down to 1 m and, being a separate sensor from the flash lidar, can provide range along a separate vector. The Laser Altimeter measurements are also fed into the ALHAT navigation filter to provide lander guidance to the safe site. The flight tests served as the culmination of the TRL 6 journey for the lidar suite and included launch from a pad situated at the NASA-Kennedy Space Center Shuttle Landing Facility (SLF) runway, a lunar-like descent trajectory from an altitude of 250m, and landing on a lunar-like hazard field of rocks, craters, hazardous slopes, and safe sites 400m down-range just off the North end of the runway. The tests both confirmed the expected performance and also revealed several challenges present in the flight-like environment which will feed into future TRL advancement of the sensors. The flash lidar identified hazards as small as 30 cm from the maximum slant range of 450 m which Morpheus could provide, however, it was occasionally susceptible to an increase in range noise due to heated air from the Morpheus rocket plume which entered its Field-of-View (FOV). The flash lidar was also susceptible to pre-triggering on dust during the HRN phase which was created during launch and transported by the wind. The Doppler Lidar provided velocity and range measurements to the expected accuracy levels yet it was also susceptible to signal degradation due to air heated by the rocket engine. The Laser Altimeter, operating with a degraded transmitter laser, also showed signal attenuation over a few seconds at a specific phase of the flight due to the heat plume generated by the rocket engine.

Execution of scientific objectives such as to investigate the central peaks of a big crater of the moon is thought to be the most scientifically interesting. Precise and safe autonomous landing ability is a productive issue for this investigation. Here it is proposed a scheme of trajectory generation for lunar descent. This research includes a sketch of a qualitative scheme on solutions of motion control equations for lunar descent vehicle from orbital speed condition down to the terminal descent situation. A trajectory generation algorithm is developed accumulating more than one step with variable thrust to mass ratio. Mathematical modeling, algorithm design, simulations and results are presented in this paper. In fact, the proposed trajectory generation scheme can readily be used to develop real-time guidance algorithm for future precise and safe lunar landing missions reducing the computational burden.

In any lunar landing mission, there comes an opportunity to demonstrate the safe and soft landing of lander on the lunar surface. The Lander module in turn carries the Rover which will touchdown the lunar surface for performing mobility and scientific operations using scientific payloads onboard. The prime mission objective of any Rover is to maximize the Rover movement over the lunar surface and operating the Rover payloads at every point and ensuring safe movement of the Rover. Chandrayaan-3. the prestigious lunar mission of India, wherein the Pragyaan Rover was launched onboard Vikram Lander which safely landed on the Lunar south pole region on 23rd August 2023 at 12:34 UT. After landing, the lander and rover initiated the targeted activities as per plan. The lander mission operations involve deploying the rover, operation of lander payloads, record and playback operations of lander and rover data. The rover mission operations include the pre-commissioning of the rover in full healthy condition, payload operations, rover mobility operations, imager operations to capture the lander and the terrain traversed by rover, data download from rover to lander, thermal, power management, final orientations for power generation of each mobility, and rover parked at an appropriate orientation conducive for power generation for next wake up etc. The first-time hands on experience of operating the rover from ground could be achieved only by appropriate mission planning and support from all the sub-system teams. All the multi-disciplinary activities were carried out semi-autonomously from ground flawlessly. This paper provides the detailed mission operations carried out at lander and rover end post successful landing.

The focus of this paper is to present an image transformation algorithm to interpret historical imagery land data for autonomous safe landing of a robotic spacecraft. Inherent geographical hazards of Martian terrain may impede safe landing for a science exploration spacecraft. Surface visualization software for hazard detection and avoidance may therefore be applied to enable an autonomous and intelligent spacecraft descent upon entering the planetary atmosphere. The methodology proposed involves integrating linear algebra and computer vision/graphics techniques, with the intrinsic parameters governing spacecraft dynamic motion and camera calibration, in order to assess hazards that might impede spacecraft landing. In this paper, we provide algorithmic details and simulation results of our methodology applied to a representative Mars landing descent profile.

In lunar landing missions, it is very important to closely estimate the horizontal velocity of a descending spacecraft for achieving a successful and safe lunar landing. The purpose of this paper is to present a novel, vision-aided approach for the accurate, efficient, and robust estimation of such horizontal velocity. Our algorithm processes images from a downward-looking camera, as well as attitude and altitude information from other sensors, to estimate horizontal velocities. During descent, images vary greatly in scale, orientation, and viewpoint. To begin, the scale-invariant feature transform (SIFT) algorithm copes with such shifts, so one is able to use extracted keypoints to establish correspondences between consecutive descent images. Then, matched SIFT keypoints are projected to the level ground plane according to the measurement of the camera state and the central projection imaging collinear equation. A 1-point random sample consensus (RANSAC) algorithm is adopted to remove mismatched keypoints. From each correctly matched keypoint pair, the algorithm derives a hypothesis for the spacecraft displacement relative to lunar ground, since those keypoints represent measurements of the same position on the lunar surface. From the bundle of displacement hypotheses, our algorithm robustly recovers the mode of the sample distribution. This final horizontal displacement estimate of the spacecraft is obtained by using the mean shift method to search for an appropriate answer among these hypotheses. In combination with the time interval between shots, the horizontal velocity is estimated. Additionally, a digital signal processor with field-programmable gate array architecture is also presented to implement velocity estimation in real time. We evaluate the performance of our algorithm based on numerous simulated image sequences and real flight images compared with the descent image motion estimation system approach and an extended Kalman filter monocular simultaneous localization and mapping method.

From June 20 through July 7 of 2009, the Autonomous Landing and Hazard Avoidance Technology (ALHAT) Exploration Technology Development Program carried out an aircraft field test over Moon like terrains to assess the use of sensors and algorithms being developed for autonomous, safe lunar landing. The field test data has been used to evaluate the performance of a lidar, a passive optical camera system, and associated algorithms for Terrain Relative Navigation. Reported here is a comprehensive description of the field test hardware, ground infrastructure and trajectory reconstruction methodologies1,2.

Future generation of landing craft will autonomously look at the surface during the terminal phase of powered descent and then, in real-time, choose and divert to a safe landing site in order to avoid hazards. Enabling technologies for such capability have been under development in recent years in the Autonomous Landing Hazard Avoidance Technology (ALHAT) project funded by NASA's Exploration Technology Development Program. ALHAT is a comprehensive system that spans the approach and landing events - from de-orbit coasting to touchdown. In this paper, we focus on ALHAT's perception task of detecting hazards in the sensed terrain and of selecting candidate safe sites for landing. This task, named Hazard Detection and Avoidance (HDA), occurs in the middle of the landing sequence. Our approach to HDA employs a probabilistic model in order to better manage the ubiquitous uncertainties associated with noisy sensor measurements and navigation. Also, we explicitly take into account the geometry of the lander and its interaction with the surface when assessing hazards. Experimental results on synthetic Lunar-like terrain show that our HDA algorithm can designate safe landing locations for a variety of terrain types and density and abundance of hazards. The complete ALHAT system is undergoing ground field-testing, and is scheduled for additional field tests on a one-hectare, lunar-like, hazard field recently constructed at NASA's Kennedy Space Center (KSC). Although the focus of ALHAT is on autonomous planetary landings, a number of terrestrial applications can also benefit from out HDA system.

The ability of lidar technology to provide three-dimensional elevation maps of the terrain, high precision distance to the ground, and approach velocity can enable safe landing of robotic and manned vehicles with a high degree of precision. Currently, NASA is developing novel lidar sensors aimed at needs of future planetary landing missions. These lidar sensors are a 3-Dimensional Imaging Flash Lidar, a Doppler Lidar, and a Laser Altimeter. The Flash Lidar is capable of generating elevation maps of the terrain that indicate hazardous features such as rocks, craters, and steep slopes. The elevation maps collected during the approach phase of a landing vehicle, at about 1 km above the ground, can be used to determine the most suitable safe landing site. The Doppler Lidar provides highly accurate ground relative velocity and distance data allowing for precision navigation to the landing site. Our Doppler lidar utilizes three laser beams pointed to different directions to measure line of sight velocities and ranges to the ground from altitudes of over 2 km. Throughout the landing trajectory starting at altitudes of about 20 km, the Laser Altimeter can provide very accurate ground relative altitude measurements that are used to improve the vehicle position knowledge obtained from the vehicle navigation system. At altitudes from approximately 15 km to 10 km, either the Laser Altimeter or the Flash Lidar can be used to generate contour maps of the terrain, identifying known surface features such as craters, to perform Terrain relative Navigation thus further reducing the vehicle s relative position error. This paper describes the operational capabilities of each lidar sensor and provides a status of their development. Keywords: Laser Remote Sensing, Laser Radar, Doppler Lidar, Flash Lidar, 3-D Imaging, Laser Altimeter, Precession Landing, Hazard Detection

After successful mission Chandrayaan - 1 India decide to launch Chandrayaan - 2. The Chandrayaan - 2 is planned to launch with a lunar orbiter, lander and rover. The lander is accomplishing autonomous soft and safe landing at the south polar region of the moon. For safe and soft landing one of the key elements is the Hazard Detection and avoidance (HDA) system which is under progress. HDA processor is process on landmarks which are based on crater topology on lunar surface from previous mission. Other parameters like matching algorithms, real time data and reference images are required for the correct navigation and exact predicted location of lander. So for storage of this reference images and real time data we required high density, space grade flash memory. For this purpose of storage RTIMS (Radiation tolerant intelligent memory stack) flash memory is used. This RTIMS flash memory controller prepared in Virtex FPGA in HDA processor. This paper presents and gives details of basic architecture, Design, comparison, VHDL implementation and test results of RTIMS flash controller.

Landing hazard detection and avoidance technology is being pursued within NASA to improve landing safety and increase access to sites of interest on the lunar surface. The performance of a hazard detection and avoidance system depends on properties of the terrain, sensor performance, algorithm design, vehicle characteristics and the overall all guidance navigation and control architecture. This paper analyzes the size of the region that must be imaged, sensor performance parameters and the impact of trajectory angle on hazard detection performance. The analysis shows that vehicle hazard tolerance is the driving parameter for hazard detection system design.