In April 2023, the superBIT telescope was lifted to the Earth’s stratosphere by a helium-filled super-pressure balloon to acquire astronomical imaging from above (99.5% of) the Earth’s atmosphere. It was launched from New Zealand and then, for 40 days, circumnavigated the globe five times at a latitude 40 to 50 degrees south. Attached to the telescope were four “drs” (Data Recovery System) capsules containing 5 TB solid state data storage, plus a gnss receiver, Iridium transmitter, and parachute. Data from the telescope were copied to these, and two were dropped over Argentina. They drifted 61 km horizontally while they descended 32 km, but we predicted their descent vectors within 2.4 km: in this location, the discrepancy appears irreducible below ∼2 km because of high speed, gusty winds and local topography. The capsules then reported their own locations within a few metres. We recovered the capsules and successfully retrieved all of superBIT’s data despite the telescope itself being later destroyed on landing.

Glacial science could benefit tremendously from autonomous robots, but previous glacial robots have had perception issues in these colorless and featureless environments, specifically with visual feature extraction. This translates to failures in visual odometry and visual navigation. Glaciologists use near-infrared imagery to reveal the underlying heterogeneous spatial structure of snow and ice, and we theorize that this hidden near-infrared structure could produce more and higher quality features than available in visible light. We took a custom camera rig to Igloo Cave at Mt. St. Helens to test our theory. The camera rig contains two identical machine vision cameras, one which was outfitted with multiple filters to see only near-infrared light. We extracted features from short video clips taken inside Igloo Cave at Mt. St. Helens, using three popular feature extractors (FAST, SIFT, and SURF). We quantified the number of features and their quality for visual navigation by comparing the resulting orientation estimates to ground truth. Our main contribution is the use of NIR longpass filters to improve the quantity and quality of visual features in icy terrain, irrespective of the feature extractor used.

In this paper we present the overall architecture of a “Science Station”, a robotically assembled and serviced persistent platform that can host multiple payloads for Earth observations. Recent decadal survey findings motivate the need to have spatial and temporal concurrency in measurements from multiple instruments. We have architected the science station to simultaneously host up to twelve Earth Venture class instruments at a time. These instruments can be replaced by newer instruments periodically to take advantage of evolving science needs and technology capabilities. The Science Station can also concurrently host science, commercial, defense and other national interest payloads. The Science Station may provide a cost-effective paradigm by mitigating some of the risks and costs associated with multiple free-flyers that may otherwise be needed for the various instruments. It leverages emergent and existent technologies in robotic assembly and servicing, lower cost commercial launch vehicles, secondary launch vehicles, and rendezvous and proximity operations. In this paper, we report the findings of a survey we conducted on the desired performance of the Science Station from various instrument hosting perspectives. We report the various trade studies that we conducted to developed a feasible architecture that meets the goals of the Science Station while also meeting the constraints of a space system. We also report the various considerations in the configuration, thermal system, pointing system, overall concept of operations, and the robotic system of the Science Station architecture. The paper then describes a testbed activity we are undertaking to evaluate the supervised autonomy robotics needed for the Science Station as well as to conduct a risk-reduction demonstration of the end-to-end robotics behaviors.

Potential future Mars Sample Return (MSR) missions could collect planetary samples and launch them into Mars orbit; in a follow-on mission, a spacecraft could rendezvous with the orbital sample (OS) to return the samples to Earth. Due to planetary protection requirements and the need to position the OS in a preferred orientation for Earth re-entry, the rendezvous phase would present a number of technical challenges. To address these challenges, this paper presents a new end-to-end testbed elements demonstrating new technologies for 1. capture of the OS within the spacecraft, 2. orientation of the OS, and 3. stowage of the OS to a Primary Containment Vessel (PCV) and internal transfer of the PCV within the spacecraft to an Earth Return Module (ERM). The end-to-end testbed consists of a 3 DOF planar robotic arm, a capture cone volume, two interchangeable orientation mechanisms, and two interchangeable internal transfer mechanisms. To simulate zero gravity during the capture stage, a cyber-physical approach is used here that fuses simulation, hardware, and autonomy elements. During the capture stage, contact dynamics of the OS with the capture cone and robotic arm end-effector is simulated using high-fidelity multibody dynamics simulation software in-the-loop. The output of the simulation is used to control the state of the physical OS in real-time using a 3 DOF robotic gantry. Additionally, the end-effector of the robotic arm is equipped with a force-torque sensor and camera to detect contact and track the OS. In the second stage, two novel mechanisms demonstrate successful orientation of the OS. In the first orientation mechanism, wipers sweep the surface of a spherical OS to engage a positive feature, thereby manipulating the OS into a preferential orientation. In the second orientation mechanism, two sets of cups selectively engage and rotate a spherical OS about two orthogonal axes; the mechanism can be operated autonomously using computer vision or interactively with a human operator in-the-loop. Finally, in the third stage, two novel mechanisms demonstrate successful internal transfer of the OS within the spacecraft volume. The internal transfer motion requires 3 DOF (rotation, translation and release of the OS). In the first internal transfer mechanism, each DOF is independently controlled using three actuators. In the second internal transfer mechanism, the 3DOF are coupled mechanically using a single actuator.

The paper addresses the problem of constructing large space structures (∼100 m) by using autonomous robots to assemble modular components in space. We are motivated by the problem of creating space structures at a scale greater than what is feasible with a single self‐deploying design. We had two goals in this work. The first was to investigate and demonstrate the feasibility of long‐order multitask autonomy. The second was to study the balance between required tolerances in hardware design and robotic autonomy. This paper reports on a payload‐centric autonomy paradigm and presents results from laboratory demonstrations of automated assembly of structures using a multilimbed robotic platform. We present results with deployable 20 lb payloads (1 m trusses) that are robotically assembled to form a 3‐m diameter kinematically closed loop structure to subcentimeter accuracy. The robot uses its limbs to deploy the stowed modular structural components, manipulate them in free space, and assemble them via dual‐arm force control. We report on results and lessons learned from multiple successful end‐to‐end in‐lab demonstrations of autonomous truss assembly with JPL's RoboSimian robot originally developed for the Defense Advanced Research Projects Agency (DARPA). Videos of these demonstrations can be seen at https://goo.gl/muNLJp (JPL, ). Each end‐to‐end run took precisely 26 min to execute with very little variance across runs. We present changes/improvements to the RoboSimian system post‐DARPA Robotics Challenge (DRC) (Karumanchi et al., ). The new architecture has been improved with a focus on scalable autonomy as opposed to semiautonomy as required at the DRC.

A potential Mars Sample Return (MSR) mission would require robotic autonomous capture and manipulation of an Orbital Sample (OS) before returning the samples to Earth. An orbiter would capture the OS, manipulate to a preferential orientation, transition it through the steps required to break-the-chain with Mars, stowing it in a containment vessel or an Earth Entry Vehicle (EEV) and providing redundant containment to the OS (for example by closing and sealing the lid of the EEV). In this paper, we discuss the trade-space of concepts generated for both the individual aspects of capture and manipulation of the OS, as well as concepts for the end-to-end system. Notably, we discuss concepts for OS capture, manipulation of the OS to orient it to a preferred configuration, and steps for transitioning the OS between different stages of manipulation, ultimately securing it in a containment vessel or Earth Entry Vehicle.

This paper presents a testbed for autonomous rendezvous and capture of small, high-speed, passive objects. While there have been several successful autonomous rendezvous and docking flight systems, a compact and standardized solution for the retrieval of very small passive objects has yet to be demonstrated. Development of a notional standard, on-orbit sample capture/return architecture would not only enable NASA missions such as potential Mars Sample Return and lunar sample return, but it could also minimize the cost and risks associated with those future missions. The testbed presented in this paper was designed for the Space Rendezvous And Capture Competition (Space RACE), a candidate competition being developed for NASA Centennial Challenges. The competition, in which competitors would build mobile robotic platforms to autonomously chase and capture a mock Orbital Sample (OS), emulates major aspects of proposed on-orbit sample capture: locating and tracking a passive target using limited fiducials, developing an approach algorithm, grappling the target, and manipulating the target for insertion into a notional Sample Return Capsule. These elements would also be pertinent to orbital debris cleanup and terrestrial applications such as warehouse packaging and autonomous harvesting.A prototype was built and successfully used to test the Space RACE concept. The prototype consists of a flat, circular track 17m in diameter, four PulsON 410 Ultra Wideband ranging radios, and a four-wheeled robotic platform called the OS-Bot. An on-board autonomous controller uses the ranging radios in the prototype to calculate the position of the OS-Bot in real-time and follow a predefined, arbitrary path. This method of navigation avoids the use of track-fixed fiducials and allows the simulation of any orbital scenario. The system also supports mounting of additional radios on competitor platforms, allowing for minimally invasive monitoring of competitor position and velocity. These aspects both prevent competitors from merely "line-following" to the OS and allow the simulation of numerous mission scenarios which approximate orbital dynamics. Results are presented in this paper for the successful testing of this platform using low resolution sensors in an environment subject to frequent multi-path ranging errors. In addition, a filter is presented which could improve the response of the controller and deliver more robust operation.

In-space assembly can enable new types of spacecraft and structures which are too large or fragile to be carried on a rocket in an assembled form, and robotic systems can make in-space assembly feasible and cost-effective. Such systems should be able to assemble large and complex structures while imposing minimal launch mass and mission risk. We propose an autonomous robotic limb, henceforth referred to as "Limbi," which is self-mobile and symmetric. Two identical electromechanical docking mechanisms serve as end-effectors. With either end-effector anchored to a base structure, the other can grab modular elements and attach them to the growing structure. Power and computing are provided by the spacecraft through these docks, enabling Limbi to walk end-over-end across the structure without a battery or tether. We have constructed and tested a prototype system in a planar workspace that demonstrates the mobility and assembly capabilities of the proposed limb. We also introduce the concept of "Limboids," consisting of multiple Limbi robots temporarily attached to each other to form more complex kinematic chains. The resulting configurations are application-specific and can be tailored to the degrees of freedom, range of motion, and general dexterity required by a particular task. Because Limbi and Limboids can assemble large and complex structures with minimal robotic complexity, the development of this class of robots is a critical step forward in low-risk and lightweight assembly.

Paul Backes, 1 Christopher McQuin, Mircea Badescu, Anthony Ganino, Harish Manohara, Youngsam Bae, Risaku Toda, Nicholas Wiltsie, Scott Moreland, Jesse Grimes-York, Phillip Walkemeyer, Eric Kulczycki, Charles Dandino, Russell Smith, Michael Williamson, Dennis Wai, Robert Bonitz, Alejandro San Martin, and Brian Wilcox Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, 91109