A robot named ASTRA, built at Caltech with funding from DARPA, has assembled a section of a truss in low Earth orbit without help from astronauts or live joystick control. Sergio Pellegrino, who leads the team, said it is the first time a structure has been autonomously constructed in space. The payload launched on March 30, 2026, and the result was announced on Nov. 5.
Tech news worth reading: Get Techpresso, the daily email with the most interesting tech news and insights, read by 400,000+ professionals from Google, Apple and OpenAI.
We may earn a commission if you sign up.
The finished piece is modest. The robot added one truss bay to two that were pre-built on the ground, and a 1.4-meter antenna frame needs 11 bays in all. The mission is scheduled to end in December 2026, and the team does not expect to finish the remaining bays before then.
ASTRA’s gantry, dispensers and 11 cameras
According to Caltech’s account of the mission, the payload is about one cubic meter and carries an armlike manipulator, motorized dispensers holding composite fiber struts and aluminum joints, and 11 cameras. The manipulator pulls parts from the dispensers and assembles identical truss bays on an assembly plate. After each bay is finished it attaches a netlike perforated film, then pushes the completed bay out into space so the cycle can repeat.
DARPA’s own description of the result frames it as a gantry robot with three linear axes joining composite truss members through custom joints while deploying a thin reflective mesh, a proof of concept for future space-based antennas. Each section is built piece by piece, given its mesh, and deployed once complete. Both accounts agree on the idea: launch compact raw material, then build the big thing where it will be used. The reflective film was co-invented by the Caltech group and collaborators at the University of Pennsylvania led by Professor Igor Bargatin. Compared with the knitted metallic mesh used on conventional furlable reflectors, it needs less prestress, which lets designers make lighter structures. The geometry behind the approach was laid out earlier in a Caltech paper, In-Space Assembly of Large Mesh Reflectors, by Jong-Eun Suh, Sahangi Dassanayake and Pellegrino.
Working through 15-minute contact windows
The robot rode to orbit on Momentus Inc.’s Vigoride 7 vehicle, which launched on SpaceX’s Transporter-16 mission from Vandenberg Space Force Base with 10 payloads aboard. Caltech says the payload was completed, tested and delivered to Momentus in under 12 months. Once in orbit, Caltech’s team sent commands through Vigoride. Launch restraints were released, the two prebuilt bays were pushed out, and the third bay was built from scratch. Commands were issued by systems engineer Charles Sommer, research engineer Alan Truong and graduate student Tara Venkatadri, with results checked against photographs downloaded from orbit. The ground crew had only about seven or eight communication windows a day, each lasting roughly 15 minutes. When something went wrong, they tried fixes on a flight spare that stayed in Pellegrino’s laboratory and then relayed the workaround to the orbiting unit. Jaden Suh, a former Caltech postdoctoral scholar now at the University of Texas at Austin, said there are always lots of little surprises. Sommer’s description of the routine was that each time an issue is confirmed in space, the team returns to the laboratory to reproduce it.
Three bays out of 11 before the December cutoff
The team ran out of time and built only the initial section, and Caltech says the mission is unlikely to complete all 11 bays before it ends in December. The institute describes the demonstration as showing that on-orbit assembly is viable and scalable to very large perimeter trusses, while noting that the design needs to be simplified and refined.
The program manager’s ambition is far larger. Andrew Detor of DARPA’s Defense Sciences Office, who says the project went from concept to a working prototype in orbit in 15 months and that operating in space is much harder than operating on the ground, said the aim of NOM4D, short for Novel Orbital and Moon Manufacturing, Materials and Mass-efficient Design, is construction at the scale of hundreds of meters or kilometers, and that nothing approaching that can be achieved today. In DARPA’s 2025 preview of the demonstrations, Detor said pushing teams into space means they cannot sweep challenges under the rug, and the agency lists 100-meter-class radio antennas and solar array farms among the long-range uses.
NOM4D began in 2022; its first two phases developed the concept and design method and then proved it in a terrestrial laboratory, and DARPA says strong phase 2 results led it to turn the final phase toward an on-orbit demonstration. ASTRA is one of two NOM4D projects to reach the program’s orbital phase. The other, from the University of Illinois Urbana-Champaign, plans to test a resin system for hardening carbon tubes on the International Space Station, as ScienceDaily’s republication of the Caltech release also notes. DARPA adds that Momentus hosted the Caltech payload, helped integrate it for launch and supplied communications, power and imaging support during the demonstration, and that NOM4D itself is winding down over the next few months.
This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.
Tech news worth reading: Get Techpresso →
We may earn a commission if you sign up.
More from Morning Overview
- A $1 billion gas plant planned for Superior, Wisconsin, has been abandoned
- Shaq spent $1.3 million on three Rolls-Royce Phantoms to prove a point to a salesman
- The Justice Department asked courts to undo approvals for three offshore wind farms, one already making power
- Over 20 trillion gallons of rain fell on the Four Corners states and Nevada in two weeks