Low-thrust propulsion systems play a crucial role in deep space exploration. However, the extended operation time required for such systems increases the likelihood of missed thrust events (MTEs), in which scheduled thrust operations cannot be executed as planned. Conventional robust trajectory design typically guarantees sufficient margin throughout the entire trajectory, which often results in overly conservative or even infeasible solutions for missions with stringent propellant budgets. We propose an MTE-probabilityaware low-thrust trajectory optimization method based on adaptivemesh sequential convex programming (SCP). By embedding a predefined distribution of MTE scenarios into the SCP framework, it simultaneously optimizes the nominal and recovery trajectories to minimize the expected propellant consumption rather than enforcing a uniform worst-case margin. The framework incorporates a fully adaptive time mesh to capture structural changes in thrust arcs and introduces a tailored step acceptance and trust-region update strategy to maintain feasibility even when these changes appear. We validate the approach in the Earth-Moon circular restricted 3-body problem (CR3BP) using a case study of an under-development secondary payload mission inserted into a direct lunar transfer trajectory. The results demonstrate that the proposed method robustifies the nominal trajectory against multiple predefined MTE scenarios with only a slight increase in propellant consumption. Despite this modest $\Delta V$ increase, the structure of thrusting arcs changes significantly: the number of arcs is reduced, and both the length and direction of the first post-separation arc, one of the most critical operations for secondary payloads, are altered. These findings highlight that probability-aware robust optimization can yield trajectories that are both propellant-efficient and operationally realistic, providing valuable initial solutions for subsequent mission design phases targeting deep space escape after a lunar re-encounter.

A water-based propulsion system is scheduled to be integrated into the GEO-X (Geospace X-ray image) mission. GEO-X mission aims to visualize the Earth's magnetosphere through X-Ray imaging and demonstrates deep space access using the spacecraft's own propulsion system. A structural thermal model (STM) of the GEO-X mission has been designed and developed in preparation for the planned launch in 2027. The STM includes the propulsion system and is used to evaluate the performance of the system. The propulsion system comprises a resistojet thruster for momentum unloading and delta-V maneuvers, as well as a gridded ion thruster for delta-V operations in deep space. Water is employed as the propellant for all thrusters and water propellant offers the advantage of liquid-phase storage, thereby facilitating a more compact tank design compared to conventional noble gas propellants. Although the reduced performance of water-based thrusters relative to noble gas propellant can result in longer orbital transfer time and labor hours associated with spacecraft operation, this mission avoids such labor hour increases by employing an autonomous operation system with the ion thruster. The autonomous system controls thruster operation by adjusting the propellant flow rate and accommodating variations in the surrounding thermal environment, in response to inputs specifying the target delta-V and orbital transfer time. Furthermore, a water circulation system is employed to support thermal management. This circulation system serves a dual purpose: it provides cooling for the water propulsion system and also functions as the water propellant source. Simulations indicate that the system achieves a thermal conductance of 0.74 W/K. The water circulation system is also planned to be incorporated as a temperature control parameter within the autonomous control system for the ion thruster.

We present a partial formal analysis of the ABZ 2026 planetary rover case study, focusing on abnormal scenarios and failure-handling behavior. From a space-systems engineering perspective, we first identified representative abnormal scenarios through what-if analysis and then formalized the relevant interactions among goal reasoning, planning, map validation, communication, and recovery-related components in Quint as a shared-state transition system. We analyzed this model through simulation and bounded verification, which revealed several specification issues and ambiguities, including a gap in the interruption semantics of safe-location instructions and underspecified decision criteria in failure handling. We also explored a focused Event-B model of the FailureMode component as a complementary analysis. The study shows how lightweight formal modeling can help domain engineers clarify abnormal behavior in an autonomous space-system specification.

This study proposes a hybrid navigation system that may achieve cost reduction while maintaining positioning accuracy for lunar navigation constellations. Currently proposed Lunar Navigation Satellite System (LNSS) concepts require all satellites to carry expensive GNSS receivers with large high-gain antennas. Our approach strategically divides the constellation into large satellites with GNSS capability and small satellites using only satellite-to-satellite tracking (SST). Through simulations evaluating GNSS+SST performance with 60 -second and 1 -second measurement intervals, we analyzed all 254 possible configurations with $1-7$ large satellites out of 8 total. Results suggest that configurations with $1-2$ large satellites can meet the 50 m position accuracy requirement under the assumed conditions, while configurations with 3 or more large satellites show performance variation depending on satellite placement. Additional evaluation of optical navigation revealed performance degradation compared to the GNSS+SST baseline under current implementation. Economic analysis using the USCM8 cost model indicates potential reductions of up to 61 % for initial deployment and 27 % for 10 -year operations compared to an all-GNSS baseline, though these estimates do not account for integration complexity or operational overhead. This analysis assumes ideal clock behavior, employs simplified simulation models, and relies on single runs per configuration rather than Monte Carlo statistics. These preliminary findings suggest that hybrid architectures may warrant further investigation as a pathway toward economically sustainable lunar positioning infrastructure.

Recent advances in space science have highlighted the demand for high-resolution astronomical and Earth observations. Achieving higher angular resolution and sensitivity requires longer focal lengths and larger apertures, which exceed the physical limits of a single satellite. Formation Flying (FF), wherein multiple satellites maintain precisely controlled relative states, offers a means to overcome these limitations and enable millimeter- to micrometer-level observation accuracy. Realizing such missions demands equally precise relative navigation and control. This study develops and validates a method for high-precision relative position and attitude determination using Quadrant Photodiode Sensors (QPS). A physics-based model of Quadrant Photodiode (QPD) outputs is constructed, and displacement estimation methods are proposed and evaluated. The approach is verified through hardware experiments and numerical simulations, which together assess the accuracy of QPS-based determination and its consistency with the developed model under realistic operating conditions.

Free-space optical communication (FSOC) is attracting significant attention as a core technology for future space communication infrastructure, owing to its capability of providing wide bandwidth and high antenna gain with narrow beam divergence. However, the small divergence angle of optical beams makes the link highly sensitive to pointing errors, requiring highly accurate tracking sensors. The quadrant detector (QD), with its simple structure and fast response, is a strong candidate for fine-tracking, but its linear response range is severely limited when placed at the focal plane. To address this limitation, this study investigates a QD defocus method in which the QD is intentionally placed out of focus to expand its field of view. Using wave-optics simulations and laboratory experiments, the response characteristics of QDs under defocus conditions are systematically evaluated. Furthermore, acquisition and tracking experiments were conducted using a numerical simulator based on the DOLCE optical terminal configuration to verify the validity of QD defocus.The results clarify the trade-off between expanded linear range and tracking accuracy, providing valuable insights for the design of compact and cost-effective optical communication terminals.

CubeSats are emerging as a promising option for deep space exploration. Despite their potential, CubeSats face significant challenges in angular momentum management due to the limited size of the reaction wheels and the minimal fuel capacity for reaction control systems. This study addresses these challenges by proposing two novel angular momentum management strategies for two major aspects of deep space CubeSats operations: cruise and trajectory correction maneuvers. Recently, an increasing number of deep space CubeSats have used rotational solar array paddles to simultaneously achieve sufficient power generation and mission observation attitude. Therefore, the first strategy involves the optimization of solar array paddle angles during cruise. Strategically adjusting the solar array paddle angles enables bias momentum control using solar radiation pressure. The second strategy focuses on robust thruster selection optimization during trajectory correction maneuvers. This method is designed to manage angular momentum while ensuring safe operations, even in the presence of propulsion anomalies. Both strategies are formulated as mathematical programming problems and validated using the deep space CubeSat EQUULEUS as a model case. The practical necessity of these approaches is highlighted through comparisons with actual on-orbit operations, and their effectiveness and broader applicability are demonstrated through numerical simulations. (c) 2025 The Author(s). Published by Elsevier B.V. on behalf of COSPAR. This is an open access article under the CC BY license (http:// creativecommons.org/licenses/by/4.0/).

Optical Satellite Networks in Low Earth Orbit enable high-capacity global communications but face unique challenges from predictable orbital dynamics and unpredictable link failures caused by cloud cover and hardware faults. While pre-computed backup routing enables fast recovery without real-time recalculation, storing complete forwarding tables for every topology snapshot and failure scenario demands excessive onboard memory and uplink bandwidth, hindering scalability. This paper presents a memory-efficient backup routing algorithm that exploits differential table compression in two dimensions. First, time-series compression stores only incremental updates between consecutive topology snapshots, since link changes due to orbital motion occur only in a limited part of the network graph. Second, backup-table compression retains only differences from nominal routes, exploiting structural similarity between failure scenarios. This approach preserves near-instantaneous reroute performance while significantly reducing storage requirements and distribution overhead. The framework integrates Link State Advertisements for both link failures and congestion, enabling dynamic traffic distribution across multiple paths without real-time computation. Numerical simulations using COSMICA demonstrate effective memory reduction while maintaining low packet loss and recovery latency on simple constellation configurations and a failure pattern.

This paper presents DOLCE (Demonstrational Laser Communication Terminal for Future Experiments), a reconfigurable ground testbed for validating acquisition and tracking control technologies in space optical communication systems. DOLCE integrates both optical and communication components, employing single-mode fiber-based free-space optical communication at 1550 nm wavelength with a target tracking accuracy of 10 mu rad. The system features two key characteristics: hardware reconfigurability through a post-based optical breadboard design, and software reconfigurability through distributed IP core architecture on FPGA. Two distinct testing configurations are supported: a tabletop test for quantitative tracking performance evaluation under controlled disturbances, and an equatorial mount test at 100-200 m distance for validating the complete acquisition-to-tracking sequence. For the equatorial mount test, we introduce a beam reducer to expand the beam divergence angle from 0.001 deg to 0.4 deg, enabling proper detection of optical axis misalignment by ensuring sufficient wavefront inclination. Additionally, we demonstrate a method for simulating satellite attitude and gimbal disturbances using the transmitter FPM, based on system identification and power spectral density synthesis. DOLCE provides a flexible validation platform for iterative development of next-generation laser communication terminals.

AQUARIUS (AQUA ResIstojet propUlsion System), a water resistojet propulsion system installed on a 6U CubeSat EQUULEUS (EQUilibriUm Lunar-Earth point 6U Spacecraft), successfully conducted multiple orbital transfers including the world's first water-fueled propulsion system operation in deep space. We present the on-orbit performance analysis of AQUARIUS with a focus on reaction control thrusters responsible for the spacecraft's angular momentum management. The reaction control thrusters were capable of producing torques about all axes for angular momentum management as intended. Additionally, they provided fine translational delta-V, demonstrating the thrusters' dual functionality and reliability in critical maneuvers. The impulse vector of each reaction control thruster was estimated by considering the thrust vector direction. The calculated impulse in the translational axis was about 0.10 N & sdot;s per shot, which was validated by comparing the impulse obtained from the Doppler shift of the communication waves. The estimated on-orbit specific impulse of the reaction control thrusters was approximately 78 s, 1.3 times larger than the measured value in the ground test. This improvement on orbit could be due to the degradation of thrust coefficient efficiency in the ground test caused by low Reynolds number flows and high background pressure lowering the apparent pressure thrust. The consumed propellant mass on orbit was estimated but deviated from the expected value under high-temperature conditions, possibly due to temperature measurement uncertainties or the influence of the on-orbit environment.

The Plasmaspheric Helium ion Observation by Enhanced New Imager in eXtreme ultraviolet (PHOENIX) onboard EQUilibriUm Lunar‐Earth point 6U Spacecraft (EQUULEUS) performed global imaging observations of the Earth's plasmasphere from a meridian view. PHOENIX is a normal‐incidence telescope designed to observe He II emission at 30.4 nm, consisting of a mirror coated with molybdenum and silicon multilayers, a thin metallic filter made of aluminum and carbon, and a microchannel plate detector. This paper provides an overview of the PHOENIX instrument, its in‐flight calibration, and initial results of Earth observations. During in‐flight calibration, it was found that stray light affected the data when the phase angle between the line of sight and the Sun was small, but a method for its removal was developed using stray light observations. The calibration results confirmed that PHOENIX is optimized for He II observation, with a sensitivity of cts/s/pix/Rayleigh for He II. It was also demonstrated that PHOENIX is capable of capturing global images of the Earth's plasmasphere with an angular resolution of less than 0.19° and a temporal resolution of less than 1.5 hr. In May 2023, PHOENIX successfully conducted imaging observations of the Earth's plasmasphere while EQUULEUS was on its way to the Earth‐Moon Lagrange point 2, revealing the density structure along the dipole‐shaped magnetic field lines. Furthermore, the shrinkage of the plasmasphere due to geomagnetic disturbances was captured. This marks the first global imaging of the Earth's plasmasphere using an ultra‐small instrument.

This paper presents the on-orbit results of AQUARIUS (AQUA ResIstojet propUlsion System), a water resistojet propulsion system installed on a 6U CubeSat EQUULEUS (EQUilibriUm Lunar-Earth point 6U Spacecraft). EQUULEUS was launched in November 2022 by NASA's SLS (Space Launch System) as one of the secondary payloads. EQUULEUS successfully conducted three delta-V operations and eleven trajectory correction maneuver operations with AQUARIUS, including simultaneous reaction control of the spacecraft, achieving a total delta-V of 18 m/s. Analysis of on-orbit data from both delta-V thrusters and reaction control thrusters confirmed that AQUARIUS was successfully utilized as designed. This demonstrates and expands the capabilities of water-based propulsion systems for small spacecraft.

The demand for space communication has been escalating due to several factors such as the expanded use of Earth observation satellites, the need for alternative communication pathways during terrestrial communication disruptions caused by natural disasters, and the necessity for robust internet connectivity in maritime and aerial environments. To meet these demands, there is a growing focus on the development of non-terrestrial networks that integrate optical communication, which offers significantly higher data rates than traditional radio frequency communication, with low Earth orbit (LEO) satellite constellations that ensure broad coverage with low latency. However, LEO optical communication constellation networks face unique challenges. These include the dynamic nature of communication links due to the orbital movement of satellites, interruptions caused by cloud cover, and the increased risk of equipment failures or malfunctions in the harsh space environment. Traditional routing methods used in terrestrial networks, when applied directly to LEO optical communication constellations, lead to frequent link switches and unexpected link failures, resulting in data loss and increased latency. This paper proposes a novel routing method to mitigate the impact of unpredictable link failures across multiple communication links, thereby minimizing packet loss and delay. For each satellite, the proposed method selects multiple communication links in the network and pre-calculates backup routing tables to be robust against the failures of the selected links. Each satellite can autonomously decide which backup table to use based on the current network status, ensuring a swift and efficient response to sudden link failures. The selection process for multiple communication links to anticipate potential failures is based on the hop count from each satellite to the link and the overall placement of ground stations within the network. This proactive strategy improves both the reliability and efficiency of LEO optical communication satellite constellations. To evaluate the effectiveness of the proposed routing method, we model the structure and data flow of a LEO optical communication satellite constellation network. This modeling facilitates precise simulation of network operations under various conditions. Numerical simulations are conducted with scenarios where sudden link failures can occur. We assess the communication performance from the simulation results and demonstrate the proposed method's capability to enhance the robustness and efficiency of LEO optical communication satellite constellation networks.