RocketSmith: An Agentic System for High-Powered Rocket Design and Manufacturing

📅 2026-05-25
📈 Citations: 0
Influential: 0
📄 PDF

career value

246K/year
🤖 AI Summary
This study addresses the lack of intelligent, integrated systems for end-to-end high-power rocket design, which hinders coordinated parameter optimization, stability validation, and manufacturability assurance. To bridge this gap, the work proposes a novel multi-agent architecture that integrates parametric modeling, flight simulation, FDM additive manufacturing, and closed-loop optimization within a unified framework supporting both zero-shot inference and human-in-the-loop interaction. The resulting system enables efficient co-design across the design–simulation–manufacturing pipeline while allowing expert intervention at any stage. Experimental validation demonstrates the successful fabrication and stable launch of four rockets with distinct configurations—two of which are reusable—with measured apogee heights deviating by only 16% from simulation predictions.
📝 Abstract
This work presents RocketSmith, an agentic system capable of the design, manufacturing, and optimization processes in high powered rocket development. The system enables the intelligent automation of software tools as to not only validate factors such as flight stability but also generate the parametric design components for the rocket assembly. A collection of subagents and skills enable optimization workflows of flight parameters via iteration in both zero-shot and human-in-the-loop workflows. With this system, four distinct high power rockets with various motor and assembly configurations were developed utilizing the unique design capabilities of additive manufacturing. These assembly components were fabricated using various FDM printers, manually evaluated for flight readiness, and flight tested at a launch event. From these tests, all rockets achieved a stable launched and two of the four rockets were successfully recovered in reflyable condition. Within the collected flight data, an 84% accuracy was achieved when comparing measured apogee to that calculated in flight simulations.
Problem

Research questions and friction points this paper is trying to address.

high-powered rocket
design automation
manufacturing optimization
flight stability
additive manufacturing
Innovation

Methods, ideas, or system contributions that make the work stand out.

agentic system
parametric design
additive manufacturing
flight optimization
human-in-the-loop
🔎 Similar Papers
No similar papers found.
P
Peter Pak
Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, PA, USA
J
Jesse Barkley
Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, PA, USA
R
Rumi Loghmani
Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, PA, USA
D
Derek Baich
Tripoli Rocketry Association, Pittsburgh, PA, USA
A
Ananya Pamal
Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, PA, USA
Amir Barati Farimani
Amir Barati Farimani
Russell V. Trader Associate Professor at Carnegie Mellon University
Computational systemsMulti-scale modelingBiophysicsDeep Learning