NASA has selected United Launch Alliance's Centaur V upper stage for the Artemis missions that aim to return astronauts to the lunar surface for the first time since 1972. The move represents a decisive abandonment of the Boeing-developed Exploration Upper Stage (EUS), a project that has become emblematic of the Artemis programme's cost control and schedule management challenges.
NASA had planned to replace the existing interim propulsion stage with the more powerful Exploration Upper Stage but that program is running behind schedule and over budget.NASA projects SLS Block 1B costs will reach approximately $5.7 billion before the system is scheduled to launch in 2028, which is $700 million more than NASA's 2023 Agency Baseline Commitment, with EUS development accounts for more than half of this cost, estimated to increase from an initial cost of $962 million in 2017 to nearly $2.8 billion through 2028.
The decision to adopt the Centaur V reflects a choice between technical ambition and fiscal discipline.NASA's filing states this approach leverages current support infrastructure and will use, with relatively minor modifications, an existing ULA upper stage, while all other alternative solutions fail to meet performance requirements, would require significant modifications to hardware still under development, or would require development of new hardware that does not currently exist, with ULA's established infrastructure representing the only currently viable opportunity for accomplishing Artemis mission objectives.The space agency will use the Centaur V, currently flying as the upper stage of ULA's Vulcan rocket, for Artemis IV and V, both slated for 2028, with a flight spare also being ordered.
The procurement decision carries institutional implications worth scrutinising.The filing formally known as a Justification for Other than Full and Open Competition allows NASA to proceed with a sole-source contract to ULA. While the technical justification appears defensible, sole-source contracting always warrants accountability.NASA noted that alternatives such as Blue Origin's New Glenn Upper Stage require significant modifications to Mobile Launcher 1, and ULA was already familiar with the steps needed to modify an upper stage for SLS. The question is whether this defence against Blue Origin reflects genuine technical constraint or merely the path of least resistance.
The stage has design similarities to the interim cryogenic propulsion stage, including the use of RL10 engines powered by liquid hydrogen and liquid oxygen, with few other upper stages having similar performance and using the same propellants as Centaur.The Centaur is a family of rocket-propelled upper stages that has been in use since 1962. This heritage matters. Proven hardware beats promising blueprints, and Isaacman's restructured Artemis timeline does not allow for technical surprises at this critical juncture.
Yet solving the upper stage equation does not solve the harder problem.The bigger challenge remains the Human Landing System as NASA still needs a version ready for a Low Earth Orbit checkout next year, with an actual lunar landing to follow in 2028. The commercial landers from SpaceX and Blue Origin remain in active development, andNASA says it is taking measures to enable the acceleration plans submitted by both HLS providers, willing to rethink requirements to achieve the objective on time and willing to make available resources and expertise to set those missions up for success.
From a national interest perspective, the move makes sense.NASA Administrator Jared Isaacman stated that NASA must standardize its approach, increase flight rate safely, and execute on the President's national space policy, with credible competition from geopolitical adversaries increasing by the day. Delays on Artemis translate to strategic vulnerability. SpaceX's rapid iteration on Starship and other nations' accelerating lunar programmes create competitive pressure that cannot be ignored.
The tighter question is whether Isaacman's accelerated cadence—targeting two lunar landings in 2028—is realistic or optimistic.Isaacman wants launches of Artemis missions to happen every ten months, while according to NASA, the average time between Apollo launches was five months. History suggests the interval is achievable in principle but demanding in practice. The present schedule leaves no margin for error on the Human Landing System or any other critical path item. Centralising that risk while stabilising the upper stage represents intellectual honesty rather than weakness, but the outcome still hinges on execution that has eluded NASA repeatedly in recent years.