BOSTON, MA, Apr 22, 2026 – Flexcompute and Northrop Grumman have developed an AI-based simulation system that models thruster plume impingement during spacecraft docking and produces results in real time. Built on NVIDIA technology, the system replaces simulation workflows that can take months and provides uncertainty estimates to support control decisions.
Plume impingement describes the interaction between a rocket thruster exhaust plume and surrounding surfaces. This interaction affects forces and thermal loads during docking operations. In vacuum conditions, gases expand rapidly, creating complex flow behavior that is difficult to replicate through ground testing. High-fidelity simulation is required to capture these effects.
Traditional approaches rely on large simulation datasets generated over extended preparation cycles. Models often require millions of simulation runs to achieve acceptable coverage and reliability. This process can take months and depends on computationally intensive physics simulations.
The Flexcompute and Northrop Grumman AI Physics model applies training methods to reduce data requirements and compute predictions. It processes simulation data alongside physical constraints and generates uncertainty estimates during inference. This approach supports control decisions while reducing mission preparation timelines by up to 100 times.
“At Northrop Grumman, we’re pioneering physics AI to accelerate design and solve complex simulation and modeling problems like plume impingement – critical for station keeping, rendezvous, and space robotics. Simply put: we’re pushing the boundaries of advanced space operations.” said Fahad Khan, director of AI foundations, Northrop Grumman. “Partnering with Flexcompute and NVIDIA, we’re accelerating innovation and mission timelines to deliver superior space capabilities for customers at the speed they need.”
Flexcompute extended the NVIDIA Physics NeMo framework for AI-based physics modeling. Flexcompute extended the framework with custom model architectures, physics-aware constraints, and training methods to nozzle plume impingement and space robotics interactions. The system integrates physics simulation outputs with AI model training and validation workflows.
“Northrop Grumman’s confidence reflects what sets Flexcompute apart,” said Vera Yang, president and co-founder of Flexcompute. “We are able to take the most accurate and scalable physics foundations and evolve them into highly trained, customized Physics AI solutions that engineers can rely on. This work shows how we are transforming the role of simulation, not just speeding it up, but expanding what engineers can confidently solve and how quickly they can act.”
“The industry’s most ambitious space missions now demand a level of speed and precision that traditional engineering cycles can no longer sustain,” said Tim Costa, vice president and general manager of computational engineering at NVIDIA. “By integrating NVIDIA PhysicsNeMo, Northrop Grumman and Flexcompute are transforming complex simulations like plume impingement from days of compute into seconds of insight, drastically accelerating the path from mission concept to orbit.”
The approach supports spacecraft operations by improving prediction speed and providing uncertainty-aware outputs for control planning. More accurate plume modeling can inform control strategies, reduce structural margins, and improve fuel use. These factors contribute to longer mission durations and efficient spacecraft operations.
Source: FlexCompute
About FlexCompute
Flexcompute is a technology company based in Madison, Wisconsin, founded in 2015 by engineers from MIT and Stanford. The company develops high-performance, cloud-native simulation software for industries including aerospace, automotive, semiconductors, consumer electronics, and renewable energy. Its key products – Flow360, Tidy3D, and PhotonForge – are designed for computational fluid dynamics (CFD), electromagnetics, and photonics simulations. These tools use GPU acceleration and AI-assisted physics modeling to enable faster simulation speeds compared to traditional methods, while maintaining accuracy. Flexcompute’s solutions aim to support hardware design and engineering workflows by reducing simulation time and improving scalability.
About NVIDIA
NVIDIA, founded in 1993 and headquartered in Santa Clara, CA, designs and manufactures graphics processing units, systems on chips, networking hardware, and AI intelligence software such as CUDA. Its products serve industries including gaming, data centers, autonomous vehicles, professional visualization, robotics, health care, and energy. The company introduced the GPU in 1999 and later expanded into accelerated computing and AI infrastructure. In gaming, its GPUs support high-performance rendering, while in AI and high-performance computing, its systems provide the infrastructure for training and deploying large-scale models. NVIDIA also develops tools for robotics and autonomous driving.
About Northrop Grumman
Northrop Grumman is an American aerospace and defense technology company that provides systems, products, and solutions in autonomous systems, cyber, C4ISR, space, strike, and logistics. Founded in 1994 through the acquisition of Grumman Corporation by Northrop Corporation, the company has a legacy in military and space technologies. Northrop Grumman serves various industries, primarily defense and government, supporting the U.S. Department of Defense, NASA, and allied nations with mission-critical systems such as B-21 Raider bombers, Global Hawk unmanned systems, and missile defense technologies. It also plays a key role in cybersecurity and intelligence solutions. As of 2024, the company employs over 90,000 people worldwide. Headquartered in Falls Church, VA, Northrop Grumman has a global footprint with engineering and manufacturing facilities across the United States and international markets.