DARPA announced that its Novel Orbital and Moon Manufacturing, Materials, and Mass-efficient Design (NOM4D) program has entered its third and final phase, after earlier phases used Earth-based materials and lunar-regolith simulants to test raw-material concepts for off-Earth manufacturing.[1] In Phase 3, Caltech will fly a free-flying experiment on Momentus’ Vigoride Orbital Services Vehicle, slated to launch on SpaceX Transporter-16 in February 2026, where an autonomous gantry robot will assemble lightweight composite fiber longerons into a 1.4-meter-diameter circular truss to mimic an antenna aperture structure.[1] The University of Illinois Urbana-Champaign team is also moving to in-space testing, partnering with Voyager Technologies for an experiment intended to launch to the International Space Station on NASA’s NG-24 mission, tentatively scheduled for April 2026.[1]
The technical significance is that NOM4D is validating the hardest missing link for lunar industry: autonomous, precision assembly of load-bearing structures from materials that are compatible with in-space manufacturing.[1][2][13] If these systems work, they reduce dependence on launch-locked monolithic hardware and support modular construction of habitats, antennae, trusses, and other large assets that are too large to launch as finished units.[1][8][13] Prior research already shows robotic in-space assembly can support large structures when designs use modular interfaces that combine mechanical, power, and data connectivity, so NOM4D is extending a maturing field from lab and simulation into orbit.[2][14]
The Ark team should monitor four things: flight reliability of autonomous assembly under orbital conditions, the performance of composite materials and connectors in vacuum and thermal cycling, the degree of onboard inspection and error-correction autonomy, and whether the demonstrations move from proof-of-concept to scalable construction economics.[1][7][14] This program is worth integrating into lunar planning because NASA and other groups are already pursuing analogous robotic assembly systems for lunar habitats and structures, meaning NOM4D may help define interface standards, robotic toolchains, and verification methods the Ark can later reuse.[8][10] The strategic question is not whether robots can assemble a truss, but whether they can become a routine, repairable, and mass-efficient construction workforce for a lunar industrial base.[1][13]