A robotic welding system designed for space could change how future missions build, repair and reuse enormous structures beyond Earth. AnalySwift LLC, a Purdue University-affiliated company, has received an $899,738 Phase II Small Business Technology Transfer award from NASA to develop technology that can weld and unweld thermoplastic composite components in space. The two-year project is aimed at solving one of the most persistent problems in lunar and Martian exploration: transporting the vast quantities of material required to establish a long-term human presence. Instead of launching every structure as a finished object, astronauts and robots could assemble large trusses in orbit or on another world, then dismantle and reconfigure them as mission needs change.
The project is being led by Kawai Kwok, an associate professor in Purdue University’s School of Aeronautics and Astronautics, who serves as principal investigator. Yu She, an assistant professor in Purdue’s Edwardson School of Industrial Engineering, will lead the robotics component. Together with AnalySwift engineers, the Purdue researchers will develop a system capable of manipulating, joining and separating composite structural components with high precision. The award follows a Phase I NASA STTR project granted to AnalySwift in 2024, which established the technical foundation for the larger effort. The Phase II work will move the concept toward laboratory demonstration, combining composite materials, embedded heating systems, robotic manipulation and structural simulation.
The central material in the project is a thermoplastic composite, a class of lightweight material increasingly considered for aerospace structures because it combines high strength with relatively low mass and can be reshaped or bonded through controlled heating. Unlike conventional thermoset composites, which undergo an irreversible chemical transformation during curing, thermoplastics can soften when heated and solidify again as they cool. That reversibility creates the possibility of welding and later unwelding structural connections. For spacecraft, the advantage could be substantial: a truss launched for one mission might eventually be separated into components, transported or rearranged, and assembled into a different structure rather than abandoned after its original purpose has ended.
The Purdue team plans to create composite joints containing resistance heaters. These heaters will be embedded within the same thermoplastic matrix used in the composite struts, alongside a thin-film resistor that converts electrical energy into heat. When activated, the heater will raise the joint to the processing temperature needed to soften the matrix at the interface between the joint and the strut. Mechanical forces can then separate the components or press them together for a new bond. The method is intended to provide localized, in situ heating rather than warming an entire structure. That precision could reduce energy use, limit thermal stress and make it possible to conduct joining operations in environments where heat must be carefully controlled.
Robots will perform the operations that would be difficult, dangerous or inefficient for astronauts to carry out manually. She’s team will develop a dual-arm robotic system equipped with vision, tactile sensing, and force and thermal feedback. Cameras will help the system identify the position and condition of each composite joint, while tactile sensors will detect contact and alignment. Force feedback will allow the robot to recognize resistance during separation or reconnection, reducing the risk of damaging a component. Thermal sensors will help regulate the embedded heaters and confirm that the joint has reached the appropriate temperature. By combining these data streams, the robot could make continuous adjustments while welding or unwelding, an essential capability when structures are large, flexible and difficult to access.
The NASA-funded activity has four connected goals: developing thermoplastic composite joints with embedded resistance heaters, building robotic manipulators for welding and unwelding, demonstrating robotic disassembly and reassembly of a prototype truss in a laboratory, and assessing whether the process could eventually be applied to full-scale structures. The laboratory demonstration will be an important test because the challenge is not simply making a strong connection. The joint must also be repeatable, controllable and capable of surviving the mechanical demands placed on a truss. A successful system would need to preserve structural integrity after multiple assembly and disassembly cycles while allowing robots to work accurately without direct human intervention.
The technology addresses a severe mismatch between the size of future space infrastructure and the dimensions of current launch systems. Communications antennas, solar arrays, thermal radiators and telescope mirrors all benefit from large supporting structures. Their performance often improves as their aperture increases, but the components needed to support them can become too large to fit inside a launch vehicle’s fairing. Current fairing diameters are limited to roughly five meters, while structures required for sustained activity around the moon, on Mars or in deep space could measure tens of meters or more. Even when a large structure can be folded or partially deployed, its mass, complexity and launch volume impose strict limits on what can be delivered from Earth.
Launch cost makes the problem even more urgent. According to AnalySwift President and CEO Allan Wood, sending payloads to low Earth orbit can cost between $2,000 and $20,000 per kilogram. Every beam, joint and support component therefore represents not only mass but also a major transportation expense. A system that allows spacecraft components to serve multiple missions could reduce the need to launch replacement structures. Large trusses might be assembled from smaller sections, expanded as new modules arrive, or dismantled and repurposed for different scientific and operational goals. In principle, the same infrastructure could support communications, power generation, thermal management or astronomical observation at different stages of a campaign.
The project also reflects a broader shift in space engineering from single-use spacecraft toward adaptable infrastructure. Traditional mission design generally treats a spacecraft or support structure as a fixed system with a limited operational life. Robotic welding and unwelding could introduce a more flexible model in which structures are treated as reconfigurable assets. A truss assembled in orbit could be modified after launch, while components damaged or made obsolete might be removed and replaced. On the moon or Mars, where transporting material from Earth will remain difficult, the ability to reuse structural elements could be particularly valuable. It could also reduce the amount of infrastructure that must be designed, packaged and launched before a mission begins.
AnalySwift will contribute its expertise in composite simulation software, which is designed to model composite structures and their behavior efficiently without sacrificing engineering accuracy. Such modeling will help researchers evaluate the strength, thermal response and reusability of the proposed joints, as well as the feasibility of applying the concept to full-scale trusses. The company’s software technologies originated at Purdue University and were licensed through the Purdue Research Foundation. If the robotic system succeeds in the laboratory, it could provide NASA with a new approach to constructing large structures in space: launch smaller components, join them where they are needed, and eventually separate and reuse them as missions evolve. That capability could become a key enabling technology for a permanent human presence beyond Earth.
Subject of Research: Robotic welding and unwelding of thermoplastic composite structures for reusable space infrastructure
Article Title: NASA-Funded Robots Could Weld, Unweld and Rebuild Giant Space Structures Beyond Earth
Web References: https://analyswift.com/ ; https://engineering.purdue.edu/AAE/people/ptProfile?resource_id=288720 ; https://engineering.purdue.edu/IE/people/ptProfile?resource_id=256650 ; https://www.purdue.edu/newsroom/2024/Q4/analyswift-receives-nasa-sttr-contract-to-transform-spacecraft-infrastructure-for-secondary-uses-during-long-duration-missions/
Image Credits: Purdue University photo/Alan Cesar
Keywords: NASA, space robotics, thermoplastic composites, robotic welding, spacecraft infrastructure, lunar missions, Mars missions, reusable structures, Purdue University, AnalySwift, space construction, composite materials
