A new milestone in aerospace engineering was achieved as LEAP 71, a Dubai-based company, successfully hot-fired its AI-designed aerospike rocket engine for 11 seconds.
Developed using the Noyron Large Computational Engineering Model, the engine burns oxygen and kerosene and delivers an impressive 1,100 pounds (5,000 N) of thrust. The aerospike engine, a departure from conventional bell-shaped designs, adapts dynamically to varying air pressure. Traditional rocket nozzles, optimized for specific altitudes, lose efficiency as atmospheric pressure decreases during ascent. The aerospike overcomes this limitation by utilizing its spike-shaped design, where combustion gases interact with the surrounding air to form a virtual nozzle that adjusts automatically to changing pressures.
This concept, though promising, has historically been difficult to perfect due to its technical complexity.
LEAP 71’s groundbreaking approach leveraged advanced AI to address these challenges. The Noyron model, trained by aerospace experts, autonomously designed the aerospike engine in just three weeks. The AI incorporated intricate physical interactions, such as thermal behaviors and performance projections, into its computations. Its iterative design process fine-tuned parameters for optimal geometry, manufacturing techniques, and projected performance.
The engine was fabricated using Selective Laser Melting (SLM), an industrial 3D printing technology, and crafted from a solid block of copper. This process allowed for precise integration of complex features, such as cryogenic cooling channels that flood the spike with liquid oxygen while kerosene cools the engine chamber. On December 18, 2024, the engine successfully endured gas temperatures of 3,500°C (6,300°F) during its first test firing in England.
Josefine Lissner, CEO and Co-Founder of LEAP 71 praised the test as a validation of their AI-driven, physics-based approach. “The spike is cooled by intricate cooling channels flooded by cryogenic oxygen, whereas the outside of the chamber is cooled by the kerosene fuel,” she explained. “I am very encouraged by the results of this test, as virtually everything on the engine was novel and untested.”
The aerospike’s success was part of LEAP 71’s ambitious “four-engines-in-four-days” testing campaign, showcasing the efficiency and innovation that computational AI can bring to aerospace engineering. While aerospike engines have existed in experimental stages since the 1950s, LEAP 71’s use of AI and 3D printing represents a transformative step toward making these engines a practical reality for space exploration.