MoonBot: Modular and On-Demand Reconfigurable Robot Toward Moon Base Construction

AI summary

Overview

Research area: Space robotics — specifically heterogeneous modular, self-reconfigurable robotic systems for lunar surface construction and infrastructure deployment.

Technical level: Advanced (hardware design engineering, modular robot architecture, and field robotics). The plain-language descriptions below make the core ideas accessible to non-specialists.

Scope: The paper presents the design, hardware development, and a preliminary lunar-analogue field demonstration of MoonBot, a heterogeneous modular robot intended to build infrastructure for a future lunar base.

What This Paper Is About

Launching mass to the Moon is extremely expensive, so rather than sending one specialized robot per task, the authors propose a set of interchangeable robotic modules that can be assembled on-site into whatever configuration a task requires. MoonBot combines articulated Limb modules, Wheel modules, Body modules, and end-effector Hand/gripper modules into morphologies such as Minimal, Vehicle, Dragon, and Multicycle. The paper's goal is to show, through real hardware and a field test in a sand-filled analogue site, that such a system can perform lunar-base-construction tasks including terrain preparation, towing infrastructure, and assisting with an inflatable habitat.

Key Contributions

1. A heterogeneous, functionality-based modular robot design (MoonBot) built from Limb, Wheel, Body, and gripper end-effector modules, with configurations including Minimal (1 × Limb + 1 × Wheel), Vehicle (1 × Limb + 2 × Wheels), Dragon (2 × Minimals connected serially), and Multicycle (multiple Minimals connected in parallel to a Body module).

2. A reported first field-test demonstration of a modular robotic system performing lunar-base-construction milestone tasks in a lunar analogue environment — which the authors argue distinguishes their work from prior conceptual-only studies of modular robots for space construction.

3. A mission-oriented task breakdown identifying three essential milestone task classes: fundamental civil engineering (rock removal, surface leveling), infrastructural hardware transport and deployment (solar tower, local communication mast), and robotic assistance in deploying a pressurized inflatable module.

4. Systematic lessons learned from field testing, with particular focus on the connector design, presented as insight for future modular robotic systems for lunar missions.

Main Findings

• Test site and scope: The field test was conducted in a 20 m × 20 m lunar analogue test site fully covered with silica sand. The authors state this is a preliminary field demonstration validating the proof of concept.

• Limb module specifications: The 7-DOF Limb is 1.55 m long, weighs 20.7 kg on Earth, can transport objects up to 2 kg, has one 1-DOF gripper at each end with a maximum opening distance of 80 mm, and uses BLDC motors with a 1:960 gearbox reduction. Power comes from two separate LiPo battery lines (11.1 V for computer and sensors, 22.2 V for actuators), allowing untethered operation for a couple of hours.

• Joint actuator specifications: Each joint actuator unit weighs 1.35 kg total, with the reduction unit weighing 1 kg. Maxon EC45 flat BLDC motors are used — 50 W versions at roll joints (θ1, θ3, θ5, θ7) and 80 W at pitch joints (θ2, θ4, θ6). The 1:960 total reduction combines a 1:160 harmonic drive and a 1:6 regular gear. The actuator delivers a constant maximum torque of 87.4 Nm and a maximum rotational speed of 5.4 rpm under full load.

• Gripper specifications: The gripper attains a maximum gripping force of 2.1 tons and operates at a velocity of 9 mm/s. A dustproof linear sliding mechanism using polyoxymethylene (POM) with minimal tolerances was chosen to limit dust infiltration, while most other parts were prototyped in polylactic acid (PLA). Gripping detection uses infrared sensors rather than physical connectors.

• Wheel module specifications: The Wheel module is 638 mm wide, has wheels 480 mm in diameter, weighs 27.2 kg, and is designed to accommodate a target payload of 30 kg. It uses a 2-DOF configuration with two independently actuated wheels, enabling differential (skid) steering.

• Body module: Prototyped in multiple 3D shapes for design exploration, including a polyhedron and a cylinder, and can carry a high-capacity battery and a high-performance on-board computer. The Body module has four grapple fixtures; the Wheel module has two.

• Reduced-DOF modular limb prototypes: A 7-DOF modular limb can be assembled from two 3-DOF units and one 1-DOF unit. Each 3-DOF unit measures 450 mm in length and weighs 10.2 kg; each 1-DOF unit measures 450 mm in length and weighs 7.2 kg. Together these give a scale consistency of 1.5 m in length with the full 7-DOF module. They use MN8017 BLDC motors, the same harmonic drive as the full limb, ODrive ODv44-ST controllers, and 14-bit resolution encoders with field-oriented control (FOC) closed-loop control.

• Connectors: Two connector types were trialed for the modular limb — a custom 3D-printed screw-type connector and a genderless diaphragm-type connector weighing 0.6 kg. The authors note that because the screw-type connector is passive, printing tolerance of the parts is very critical for proper attachment.

• Field demonstration tasks: A Dragon configuration towed a deployable solar tower structure weighing over 30 kg by using its rear Limb as a fixed manipulator gripping a sledge. A team of a Minimal and a Dragon configuration cleared rocks and raked the field as civil engineering preparation. MoonBot also assisted inflatable-module deployment by monitoring inflation status and inserting rolling-proof stoppers after full inflation.

• Motivation for heterogeneity: The authors argue that while dividing a system into smaller modules increases flexibility, it also raises the number of inter-module connections and the complexity of reconfiguration operations, introducing additional potential failure points. They therefore adopt a heterogeneous, functionality-based design to balance modularity against operational simplicity, since robot operable time on the Moon is severely limited.

• Dragon configuration rationale: The paper states that connecting two Minimal units into a Dragon increases ground reactive forces, described as a critical factor under lunar gravity (1/6 G).

• Not reported in the available content: The truncated paper content does not include the quantitative results of the field demonstrations, success/failure rates, detailed lessons-learned findings, or the teleoperation control software stack results that the abstract and introduction promise in later sections.

Methodology in Plain English

The team took a mission-first approach: instead of designing one robot for one job, they built a family of module types and let the required task determine which modules get combined.

The hardware was developed as a ground-testing prototype rather than a flight-qualified system. Structural housings were made by stereolithography (SLA) 3D printing in EPX82 epoxy-based resin, with internal mechanical structures milled from duralumin, and the limb geometry was refined using finite element analysis (FEA) to find high-stress areas and reduce unnecessary material. Predicted structures were then checked with manual loading tests. Joint actuators used super duralumin and carbon-infused resin for low weight.

Each module is a self-contained system with its own power, computing, sensors, and communication. The Limb's primary control unit is a LattePanda Alpha DFR0547 board running Ubuntu 20.04 LTS; the reduced-DOF modular limb uses a LattePanda 3 Delta communicating with ODrive controllers over CAN bus. Sensing includes motor current sensors, Hall-effect joint angle sensors, photo-reflective sensors for joint zero position, an IMU, and battery voltage gauges. Teleoperation is done over Wi-Fi via an antenna on the link casing.

For the field validation, the researchers built configurations out of real modules and had them execute task scenarios in a sand-filled analogue site: assembling modules on a launch-lock palette, forming Minimal and Dragon configurations, clearing and raking terrain, towing a solar tower structure on a sledge, and supporting the deployment of an inflatable module. They then distilled lessons learned, with emphasis on the connector design.

Why This Matters

Impact on research: The paper positions itself against prior modular-robotics-for-space work that stayed conceptual, offering what the authors describe as the first field testing report demonstrating a heterogeneous modular system performing lunar base construction milestone tasks. It also contributes design-level detail — actuator sizing, reduction ratios, connector trade-offs, dust mitigation — that is often absent from conceptual studies.

Real-world applications:

• Robotic preparation and construction of lunar base infrastructure, including solar power stations, communication masts, and shelters.
• In-situ assembly of large structures from modules launched compactly, in the spirit of how the International Space Station was assembled (referenced in the paper).
• Resource-utilization and exploration missions in the lunar polar region, where permanently shadowed areas are considered resource-rich and where sunlight arrives at low horizontal angles.
• More broadly, modular serviceable robotics where a failed module can be replaced rather than the whole system, reducing downtime and maintenance cost.

Industry relevance: The work speaks to launch-mass economics — the paper notes transport costs often exceeding millions of dollars per kilogram — and to the sustainability of long-term space operations. For the space robotics and aerospace industry, it offers a concrete argument for heterogeneous modular architectures and a set of engineering parameters (mass, torque, payload, reduction ratios) that can inform future platform development. The paper's stated timeline places infrastructure establishment in the 2040s, a permanent lunar base by the 2050s, with initial deployment of selected modules anticipated in the 2030s.

Future Directions

• Environmental qualification: The paper explicitly states that launch and landing vibration resistance, dust contamination, thermal cycling, and radiation were not fully addressed. Planned future work includes vibration experiments, dust-proof evaluations, thermal vacuum testing in environmental chambers, and radiation exposure assessments in collaboration with specialized facilities.
• A first lunar mission: The authors envision launching a minimal number of modules to the Moon to validate in-situ self-assembly, reconfiguration, locomotion, and manipulation, with additional modules deployed afterward to progressively undertake actual lunar infrastructure construction.
• Connector refinement: The lessons learned focus on connector design, and the modular limb work references an improved diaphragm-type connector with an error-absorption mechanism, indicating connector robustness is an ongoing open problem for passive, 3D-printed interfaces.
• Simplified limb designs: The paper states that future iterations of MoonBot are actively investigating lower-DOF limb designs for simplified assemblies, since in many cases rotational actuation of a wheel's vertical axis is sufficient and additional degrees of freedom are redundant.

Target Audience

This paper is most useful to space robotics researchers and engineers working on modular, reconfigurable, and self-assembling robotic systems; lunar mission architects and planetary infrastructure planners; and robotics groups interested in hardware design trade-offs for extreme environments. Engineers focused on actuator design, dust-tolerant mechanisms, or inter-module connectors will find the specification-level detail valuable. Because the paper is a hardware-and-field-test report rather than a control-theory or machine-learning contribution, readers seeking algorithmic results will find it less directly applicable.

Authors’ abstract

The allure of lunar surface exploration and development has recently captured widespread global attention. Robots have proved to be indispensable for exploring uncharted terrains, uncovering and leveraging local resources, and facilitating the construction of future human habitats. In this article, we introduce the modular and on-demand reconfigurable robot (MoonBot), a modular and reconfigurable robotic system engineered to maximize functionality while operating within the stringent mass constraints of lunar payloads and adapting to varying environmental conditions and task requirements. This article details the design and development of MoonBot and presents a preliminary field demonstration that validates the proof of concept through the execution of milestone tasks simulating the establishment of lunar infrastructure. These tasks include essential civil engineering operations, infrastructural component transportation and deployment, and assistive operations with inflatable modules. Furthermore, we systematically summarize the lessons learned during testing, focusing on the connector design and providing valuable insights for the advancement of modular robotic systems in future lunar missions.