Hundreds of pits on the Moon and Mars expose unique unweathered geology on their walls, and some may access habitable caves. Small, autonomous rovers could explore, image, and map planetary pits from the rim, capturing close-range, low-angle, long-exposure imagery that cannot be captured from orbit. The small but high fidelity models must be processed in-situ for the short-duration small missions of our time with limited bandwidth since the raw imagery required for modeling is at least an order of magnitude greater than what can be downloaded in a lunar day. The vast raw imagery is the basis for modeling pits with the accuracy, resolution and lighting corrections necessary for morphology study and search for cavernous openings. This work develops and evaluates several component technologies for these explorations. A safe-approach behavior is developed and employed to navigate pit rims to vantages for acquiring images. Images from multiple vantages are fed to a specialized incremental pit modeling pipeline that computes the high-resolution 3D pit models. The technologies are manifested aboard a prototype micro rover and demonstrated at a terrestrial pit having size and shape comparable to known planetary pits. In multiple end-to-end mission simulations, these techniques are shown to reliably produce unprecedented, accurate, high-coverage pit models.

Nearly half a century ago, two papers postulated the likelihood of lunar lava tube caves using mathematical models. Today, armed with an array of orbiting and fly-by satellites and survey instrumentation, we have now acquired cave data across our solar system-including the identification of potential cave entrances on the Moon, Mars, and at least nine other planetary bodies. These discoveries gave rise to the study of planetary caves. To help advance this field, we leveraged the expertise of an interdisciplinary group to identify a strategy to explore caves beyond Earth. Focusing primarily on astrobiology, the cave environment, geology, robotics, instrumentation, and human exploration, our goal was to produce a framework to guide this subdiscipline through at least the next decade. To do this, we first assembled a list of 198 science and engineering questions. Then, through a series of social surveys, 114 scientists and engineers winnowed down the list to the top 53 highest priority questions. This exercise resulted in identifying emerging and crucial research areas that require robust development to ultimately support a robotic mission to a planetary cave-principally the Moon and/or Mars. With the necessary financial investment and institutional support, the research and technological development required to achieve these necessary advancements over the next decade are attainable. Subsequently, we will be positioned to robotically examine lunar caves and search for evidence of life within Martian caves; in turn, this will set the stage for human exploration and potential habitation of both the lunar and Martian subsurface.

Lunar pits are windows to great unknowns beneath the surface of the Moon. The steep pit walls expose a vertical cross-section of stratified lunar crust, a natural historical record of the Moon's formation. The shadowed pit floors conceal possible entrances to caves or lava tube networks where astronauts might shelter from the hazards of the lunar surface. While orbital imagery has led to the discovery of more than 200 lunar pits, surface exploration is needed to capture the close-range, long exposures required to address questions of lunar formation and habitation. We envision an autonomous micro-rover for rapid surface exploration and in situ modeling of a lunar pit. We survey the specific autonomy, mapping, and roving technologies required, and we demonstrate the accuracy and coverage achievable by pit modeling in an evaluation performed at a terrestrial pit.

To the Editor-2021 is the International Year of Caves and Karst (IYCK).To honor this occasion, we wish to emphasize the vast potential embodied in planetary

Lunar ice, present at the poles, could be a source of water for drinking and growing crops, generating oxygen for breathing, and producing propellants for venturing beyond the moon to deep space. Viability of this grand vision depends on specifics of the accessibility, depth and concentration of the ice, which can only be determined by surface missions. This requires a campaign of diverse, repeated robotic explorations over time. This paper profiles MoonRanger, a micro-rover manifested on a 2022 NASA Commercial Lander Payload Services (CLPS) flight, which will be the first such mission, with its technologies, designs and operations for a pioneering in-situ measurement of lunar ice. The paper addresses the challenges of and solutions for micro-rover perception in darkness, navigation and mapping autonomy with limited computing, mechatronic miniaturization, mobility in polar terrain, thermal management, power generation, and others. MoonRanger's mission principles, designs, analyses and performance of the innovations, component technologies, rover, system and mission. Details covering rover design, its performance, and system elements such as stow-deploy, power, avionics, thermal, software and communication scenarios are articulated. The paper describes principles, instrument, and means for measuring ice. System considerations such as ConOps, ground control, and risk posture are addressed. The paper touches on the creative, operational, lean and resourceful development culture that is producing this. The paper concludes by discussing the "firsts", the impacts of the technical contributions, the status of the development, and the path to mission.

Lunar pits are windows to great unknowns beneath the surface of the Moon. The steep pit walls expose a vertical cross-section of stratified lunar crust, a natural historical record of the Moon’s formation. The shadowed pit floors conceal possible entrances to caves or lava tube networks where astronauts might shelter from the hazards of the lunar surface. While orbital imagery has led to the discovery of more than 200 lunar pits, surface exploration is needed to capture the close-range, long exposures required to address questions of lunar formation and habitation. We envision an autonomous microrover for rapid surface exploration and ​in situ modeling of a lunar pit. We survey the specific autonomy, mapping, and roving technologies required, and we demonstrate the accuracy and coverage achievable by pit modeling in an evaluation performed at a terrestrial pit. Figure 1: Artist’s depiction of a pit exploration rover imaging the opposing wall from a vantage on the rim.

Micro-rovers offer immense advantages of low mass, low cost and frequent flight opportunities. Due to the constraint of low mass micro-rovers of our time cannot be isotope-heated. Therefore, they cannot survive the extended planetary nights, so they must achieve their exploration goals in a single daylight period. Their small size, mass and power precludes a radio for direct communication with Earth. For this reason, they can only receive and relay data while in proximity to their lander, and hence, they cannot be constantly supervised or teleoperated from Earth like larger rovers with greater power and communication capability. In order to explore beyond lander communication range, micro-rovers must operate autonomously. Micro-rover autonomy software must achieve communication-denied, high-cadence, kilometer-scale exploration treks. This paper formulates a software architecture and component-wise design for achieving the required autonomous micro-rover exploration. This technology will be integral to the MoonRanger micro-rover, which will fly to the lunar pole in December 2022 as a Lunar Surface Instrument and Technology Payload (LSITP) aboard the Masten XL1 lander. MoonRanger will conduct long treks from and to the lander to explore for lunar polar ice. The software will incorporate perception, planning, navigation, and execution, log data. Upon return to its lander, it will transfer data, images and scientific information that are result from mission-relevant autonomy. It will do so at a leap of performance beyond that achieved in prior planetary roving, but with the power, sensing and size constraints of microroving. MoonRanger hosts a two-computer system consisting of a space-hardened embedded processor and a higher-performance, less-hardened computer. Autonomy software and image processing run on a Linux-based OS on the higher-performance computer, while motor control, sensor data collection, and low-level functionality run on a real-time OS aboard the embedded processor. A design prototype of the higher-performance computer’s software is depicted in Figure 1. As shown in the figure, this software is organized into two categories, the navigation pipeline and the execution nodes. The navigation pipeline performs perception, rover pose estimation and planning, while the execution nodes handle executive control, data management and transfer, health monitoring and telemetry management. Figure 1: Software Architecture. The navigation pipeline is depicted in violet. Execution nodes are depicted in blue. The Global Planner will run on ground software. The lander, cameras, embedded processor, and disk are external to the design.

Summary: This paper details the first complete, DOE approved, deployed and operational radiometry for robotic nondestructive assay (NDA) of holdup deposits in gaseous diffusion piping. Some features, like peak-finding, Compton correction, detector efficiency, and material properties are characteristic of typical manual methods. Most features are specific to in-pipe robotic deployment. These include in-motion radiometry, self-determined location, geometric modeling of thick deposits, bounding of self-attenuation thickness and auto-checking of replicate measurements. Significant are the means for determining all the properties, equations, and constants for automatically computing and displaying all the quantities, uncertainties and information needed for NDA reporting, analysis, and review. Earlier work by this team demonstrated rudimentary but convincing robotic in-pipe NDA of holdup deposits in pipes. Cold testing with high-enrichment mat sources and hot tests with low U-235 loadings together succeeded to exhibit convincing proof-of-principle. This research-grade radiometry implementation lacked accurate odometry, efficiency calibration, high-fidelity detector modeling, attenuation modeling, forward geometric modeling, auto-segmenting, uncertainty modeling, replicate checking, deconvolution and much more. This work explains the technical basis behind robotic in-pipe NDA. It includes discussion of a new method for modeling of self-attenuation using in-pipe deposit geometry information, detail of the method’s total measurement uncertainty, and explanation of calibration methods and results. Beyond development of automation-specific radiometric elements, the paper highlights analyses, testing, and technical bases specific to and supportive of the methodology. Examples include collimation characterization, detector efficiency determination, auto-calibration methodology, and odometric replicate-checking. Results of this methodology are formalized in a DOE EM Technical Basis Document, implemented in software, formally verified in review and acceptance testing and proceduralized as a means for D&D NDA at the Portsmouth enrichment facility. Application/benefit to others: Measurement of U-235 quantity is a well-known cost and schedule driver in D&D of every defunct gaseous diffusion enrichment facility in the world. The radiometry methodologies and results detailed in this paper were the high hurdles. Having established the technical basis for this automated NDA, the opportunity is to apply this to many sizes and robot forms for addressing holdup deposit NDA throughout the D&D community. INTRODUCTION A significant challenge in achieving “criticality incredible (CI)” status for D&D of gaseous diffusion facilities is characterization of every pipe and component to ensure that the amount of U-235 in holdup deposits is below the CI threshold. Any piping that exceeds the criterion incurs costly removal, handling, and management before disposal. Pipes below the criterion are candidates for economical demolishing in place without severing, lowering, handling, or cleaning. The current method of assay involves manually positioning radiation detectors on pipe exteriors for miles of pipe [1]. This requires extended work at elevation and removal of enclosure panels and other obstructions. What follows is lengthy transcription, analysis, reporting, and archiving, which are largely WM2019 Conference, March 3 – 7, 2019, Phoenix, Arizona, USA manual and often lengthy processes. These manually-deployed assays and their analysis/reporting processes are significant and well-known cost drivers and schedule bottlenecks within DOE D&D. Chartered by DOE at the Portsmouth Gas Diffusion Plant in 2017, the Pipe Crawling Activity Measurement System (PCAMS) is an in-pipe robotic and auto-analysis package for automatic NDA of defunct gas diffusion piping. Deployment of the system has demonstrated faster, more certain NDA results while minimizing risk to personnel and need for human intervention. Deployed in research form in late 2017 [2,3], PCAMS is now a production prototype delivery successfully hot tested in July 2018 and slated to begin operations in January 2019. Initial deployment of 30and 42-inch pipe inspection is projected to cover 50,000 feet of piping and with huge positive beneficial cost and schedule impacts in two process buildings at the Portsmouth site alone. Advancements in method, analysis, and robotic technology, proceduralization and operations are currently being applied to development of smaller-size pipe inspection (3and 10inch) for other potential facilities including the sister diffusion plant in Paducah, Kentucky. The enabling advantage of the PCAMS method is traversal along and direct observation of contaminated pipe wall interiors (Fig. 1). Interior inspection allows the “RadPiper” robot to travel unobstructed while carrying a novel disc-collimated detector which exploits the axisymmetry of piping to speed assay and simplify calibration, analysis, and qualification of method. This automation is in contrast to manual methods’ need to deploy personnel in full PPE at elevation to remove miles of protective paneling and directly access the exterior of each foot of process piping over which to dwell detectors. These methods require extensive plant disassembly and dwell time, up to fifteen minutes per five feet of pipe. RadPiper instead automatically traverses piping at 10 feet (for 30-inch pipe) or 6 feet (for 42-inch) per minute and autonomously reaches its safe endpoint before reversing to its origin, collecting replicate data on reverse. Fig. 1. (Left) PCAMS RadPiper traversing a cutaway pipe and (right) deploying into a process pipe. Fifty thousand feet of the pipe immediately assayable by PCAMS at DOE Portsmouth is contained within two large process buildings, the challenge of which is two-fold. First, the initial set of pipes to inspect are large diameter: 30and 42-inch. These large diameters mean that even small percentages of pipe wall can include large amounts of holdup deposit. Second, this section of the cascade handled low-enriched uranium, meaning that deposits of the same U-235 content are much thicker and much more highly selfattenuated than at higher enrichments. The initial RadPiper robot and PCAMS method are designed to handle these particular challenges, with extensive self-attenuation modeling and geometric inspection of pipe surfaces. Compliance with DOE EM standards [4,5] is discussed further in [6]. WM2019 Conference, March 3 – 7, 2019, Phoenix, Arizona, USA The full PCAMS radiometric method is summarized in Eq. 1, which reports mass of U-235 per length of pipe. Reading left to right, the 186 keV count rate term is conservatively adjusted based on the characteristic smoothing of disc-collimated detector assay . The self-attenuation factor further adjusts this result based on PCAMS novel calculation method, which takes advantage of direct geometric profiling of deposit thickness. The 186 keV activity of U-235 combines with the calibration efficiency to determine a characteristic count rate per gram-per-foot conversion factor. The final term conservatively adjusts the reporting from a per-field-of-view to per-foot basis for NDA and CI determination .

To decommission deactivated gaseous diffusion enrichment facilities, miles of contaminated pipe must be measured. The current method requires thousands of manual measurements, repeated manual data transcription, and months of manual analysis. The Pipe Crawling Activity Measurement System (PCAMS), developed by Carnegie Mellon University and in commissioning for use at the DOE Portsmouth Gaseous Diffusion Enrichment Facility, uses a robot to measure Uranium-235 from inside pipes and automatically log the data. Radiation measurements, as well as imagery, geometric modeling, and precise measurement positioning data are digitally transferred to the PCAMS server. On the server, data can be automatically processed in minutes and summarized for analyst review. Measurement reports are auto-generated with the push of a button. A database specially-configured to hold heterogeneous data such as spectra, images, and robot trajectories serves as archive. This paper outlines the features and design of the PCAMS Post-Processing Software, currently in commissioning for use at the Portsmouth Gaseous Diffusion Enrichment Facility. The analysis process, the analyst interface to the system, and the content of auto-generated reports are each described. Example pipe-interior geometric surface models, illustration of how key report features apply in operational runs, and user feedback are discussed.

Miles of contaminated pipe must be measured, foot by foot, as part of the decommissioning effort at deactivated gaseous diffusion enrichment facilities. The current method requires cutting away asbestos-lined thermal enclosures and performing repeated, elevated operations to manually measure pipe from the outside. The RadPiper robot, part of the Pipe Crawling Activity Measurement System (PCAMS) developed by Carnegie Mellon University and commissioned for use at the DOE Portsmouth Gaseous Diffusion Enrichment Facility, automatically measures U-235 in pipes from the inside. This improves certainty, increases safety, and greatly reduces measurement time. The heart of the RadPiper robot is a sodium iodide scintillation detector in an innovative disc-collimated assembly. By measuring from inside pipes, the robot significantly increases its count rate relative to external through-pipe measurements. The robot also provides imagery, models interior pipe geometry, and precisely measures distance in order to localize radiation measurements. Data collected by this system provides insight into pipe interiors that is simply not possible from exterior measurements, all while keeping operators safer. This paper describes the technical details of the PCAMS RadPiper robot. Key features for this robot include precision distance measurement, in-pipe obstacle detection, ability to transform for two pipe sizes, and robustness in autonomous operation. Test results demonstrating the robot's functionality are presented, including deployment tolerance tests, safeguarding tests, and localization tests. Integrated robot tests are also shown.

Current methods for inspection of spent nuclear fuel storage basins involve lowering a single camera for visual inspection of walls and other structures. We present a localized inspection solution where the images are automatically annotated by localization information and a 3D model of the inspected area is generated. The system consists of an underwater sensor pod containing a stereo pair of cameras, light source, inertial measurement unit, and a pressure sensor. The sensors are time synchronized to provide precise measurements. We describe both the sensor pod and the algorithms that keep the pod localized. Preliminary results from in-air and underwater testing of a prototype are presented.

Future robotic missions to the poles of the Moon and Mercury will face challenges not encountered by current and prior planetary rover missions. Careful energy-aware spatiotemporal path planning will be required to accomplish mission objectives at high cadence under changing illumination conditions. With attention to landing site and time, such spatiotemporal path planning may enable extended missions on order of months that would not otherwise be possible. This work presents improvements in energy-aware spatiotemporal path planning for multiple waypoints to significantly reduce planning time. A time-compression technique is used to simplify planning in areas where changes occur infrequently. Consideration of end-goal reachability reduces the search space. Finally, heuristics that use pre-computation with static obstacles speed up the search.

Robots will be the first to discover and characterize ices that exist at the poles of some moons and planets. These distinctive regions have extensive, grazing, time-varying shadows that raise significant time and energy constraints for solarpowered robots. In order to maximize the sciencevalue of missions in such environments, rovers must visit as many targets as possible while considering limitations imposed by time-varying shadows and risks associated with traveling long distances. This paper compares a greedy baseline algorithm with two genetic algorithm approaches for selecting and sequencing waypoints to maximize waypoint value while minimizing distance traveled. The value and diversity of solutions from the baseline greedy solution, a single-objective genetic algorithm, and an NSGA-II framework are compared for this multiobjective optimization problem. All genetic solutions are shown to find high value sequences as compared to the greedy algorithm. This research demonstrates that a genetic approach could be utilized to effectively plan future missions for solar-powered rovers in dynamic, shadowed environments.

This paper addresses the problem of planning views for modeling large, local, substantially 3D terrain features at long range from surface rovers. These include building-size and stadium size pits with vertical walls. Pits have been identified in recent high-resolution images of the Moon and Mars. Planetary pits are interesting scientific targets created by collapse, often exposing layers of bare rock in their walls, hinting at past volcanism and other subsurface processes with their morphology. Some offer glimpses into caves. This paper presents a pipeline for view trajectory planning that enables detailed modeling of planetary pits from surface rovers. Techniques for converting prior terrain knowledge into a planning problem are developed, methods for planning rover images are discussed, and a comparison of different image-based reconstruction methods for pit modeling is presented. Results from preliminary field experiments for the end-to-end view trajectory planning pipeline are presented.

This paper presents a method that applies connected component analysis to plan routes that keep robots continuously illuminated and on traversable slopes while reaching one or more goal locations. Such routes promise to extend the lifespan, range, and scientific return of solar-powered robots exploring environments with changing but predictable lighting conditions, particularly those of the Moon and Mercury. Maps of lighting and ground slope that describe these constraints in position and time are computed, and all distinct interconnected regions that have both direct sunlight and safe slope are found using connected component analysis. These three-dimensional connected components are pruned of roots that violate time constraints and branches that dead-end in discontinuous routes. Each component is the basis for a graph that includes all feasible routes from the initial time to the final time of that component. The shortest feasible route between a pair of start and goal positions within the same component is found using A* search and is characterized by its total length and average speed. Malapert Peak and Shackleton Crater, both near the Moon's South Pole, serve as examples throughout this paper due to their highly-relevant, dynamic, and predictable lighting caused by the Moon's motion relative to the Sun.

Future planetary robotics will require path planning that ensures rover safety while responding nimbly to new information and evolving goals. Such missions include prospecting for ice at the lunar poles. These challenge robotic explorers in ways not encountered in equatorial missions that benefit from high solar elevations. Longdistance traverses must be achieved over short timescales. The grazing sun angles cause substantial, time-varying shadows that require paths to consider rover temperature and power balance in addition to terrainability. This paper details a hierarchical planner that considers these elements to rapidly generate long-distance paths. The planner has two components: a high-resolution planner that pre-computes feasible trajectories over short distances and a low-resolution mission planner that leverages the results of the kinematic planner to quickly find long-distance, feasible paths. Planner capabilities are demonstrated in simulated traverses on the poles of the Moon.

While planetary pits and caves have been fiction for a century, they have been seen from orbit only in the last few years. These discoveries exceed the fantasies in diversity, scale, and abundance. For pits and caves, this is the age of discovery, ranging from a few pits on the Moon and Mars in 2009 to hundreds within the time of this research, with many more to come. Pits with subsurface voids have been confirmed on the Moon and Mars and indicated on Venus, Phobos, Eros, Gaspra, Ida, Enceladus, and Europa. Compelling next steps are surface and subsurface exploration.Pits and caves are opportunistic study targets for unique origins, geology, and climate that will broadly impact planetary science. Holes on Mars are of particular interest because their interior caves are relatively protected from the harsh surface, making them good candidates to contain Martian life. Pits are prime targets for possible future spacecraft, robots, and even human interplanetary explorers. Caves and caverns could be ready-_made shelters for future Moon and Mars explorers and colonists. Discoveries to date look down from on high with satellites but cannot reveal the wonders of caves. They cannot enter, touch, or view pits up close. Genuine exploration is only achievable through surface missions. Robotic missions can assess suitability for safe entry and habitation, plus inform techniques for developing subsurface infrastructure.Missions into planetary voids redefine the future of exploration, science, and habitation beyond Earth. We can reach this future only by targeting specific technological advancement now. Prior missions and current roadmap priorities target regions of benign terrain. While in-cave concepts have been postulated, the critical technologies have not been identified and demonstrated.While robotic exploration of skylights and caves can seek out life, investigate geology and origins, and open the subsurface of other worlds to humankind, it is a daunting venture. Planetary voids present perilous terrain requiring innovative technologies for access, exploration, and modeling. These same technologies are broadly applicable to explorations of rough and/or subsurface planetary environments, including caves, craters, cliffs, and rock fields. This research speculates on the possibilities and means of such exploration with fundamental contributions to exploring, modeling, and visualizing this new class of large-scale, highly three-dimensional concave planetary features.

Caves on other planetary bodies offer sheltered habitat for future human explorers and numerous clues to a planet's past for scientists. While recent orbital imagery provides exciting new details about cave entrances on the Moon and Mars, the interiors of these caves are still unknown and not observable from orbit. Multi-robot teams offer unique solutions for exploration and modeling subsurface voids during precursor missions. Robot teams that are diverse in terms of size, mobility, sensing, and capability can provide great advantages, but this diversity, coupled with inherently distinct low-level behavior architectures, makes coordination a challenge. This paper presents a framework that consists of an autonomous frontier and capability-based task generator, a distributed market-based strategy for coordinating and allocating tasks to the different team members, and a communication paradigm for seamless interaction between the different robots in the system. Robots have different sensors, (in the representative robot team used for testing: 2D mapping sensors, 3D modeling sensors, or no exteroceptive sensors), and varying levels of mobility. Tasks are generated to explore, model, and take science samples. Based on an individual robot's capability and associated cost for executing a generated task, a robot is autonomously selected for task execution. The robots create coarse online maps and store collected data for high resolution offline modeling. The coordination approach has been field tested at a mock cave site with highly-unstructured natural terrain, as well as an outdoor patio area. Initial results are promising for applicability of the proposed multi-robot framework to exploration and modeling of planetary caves.