Firefly Aerospace has announced a strategic collaboration with NVIDIA to bring advanced edge AI processing to lunar orbit. According to the company’s press release, this partnership will power “Ocula,” billed as the first commercial lunar imaging and mapping service.

The Ocula service will be deployed on Firefly’s Elytra orbital vehicle during the upcoming Blue Ghost Mission 2, which is targeted for launch no earlier than late 2026. By integrating an NVIDIA Jetson module with Firefly’s proprietary SciTec AI software, the spacecraft will process high-resolution images directly in space.

This on-orbit processing capability aims to bypass traditional deep-space communication bottlenecks, delivering real-time, actionable insights back to Earth rather than transmitting raw, bandwidth-heavy data files.

The Ocula Service and Mission Profile

Hardware and Software Integration

The technological core of the Ocula service relies on high-resolution telescopes provided by the Lawrence Livermore National Laboratory (LLNL). Embedded directly into these telescopes is the NVIDIA Jetson edge AI module, which serves as the primary processing engine for the spacecraft’s optical sensors.

Driving the hardware is AI software developed by SciTec, a subsidiary of Firefly Aerospace. The company notes in its release that these algorithms have already been proven in critical national security missions in Earth orbit, providing a reliable foundation for their deployment in deep space.

Mission Timeline and Expansion

Elytra will initially serve as a transfer vehicle and long-haul communications relay for Firefly’s Blue Ghost lunar lander during Blue Ghost Mission 2. Following these initial duties, Elytra will remain in lunar orbit for approximately five years to operate the Ocula service.

Firefly is already looking beyond this initial deployment. The company is under contract to deploy two additional Elytra vehicles during Blue Ghost Mission 3 and Mission 4. This constellation approach is designed to increase coverage and reduce revisit times over the lunar surface.

Overcoming Deep Space Bottlenecks

Spacecraft equipped with high-resolution optical sensors generate massive volumes of raw data. Historically, transmitting this data back to Earth has been severely hindered by the latency and limited bandwidth of deep-space communication networks.

To solve this issue, the aerospace industry is shifting toward edge computing. By processing data on the spacecraft using AI, the vehicle can analyze raw imagery autonomously and only transmit the most important insights or compressed data back to Earth.

“Modern space missions generate massive volumes of data that require immediate processing to overcome the latency and bandwidth constraints of deep-space communications,” stated Deepu Talla, VP of Robotics and Edge AI at NVIDIA, in the press release. “Integrating the NVIDIA Jetson platform into Firefly’s Elytra spacecraft enables autonomous, on-orbit AI processing that transforms raw lunar imagery into actionable insights in real time.”

Dual-Use Capabilities for Commercial and Defense Sectors

Lunar Mapping and Space Domain Awareness

Ocula is positioned as a dual-use service catering to both commercial and government customers. Its primary capabilities include continuous, high-resolution imaging of the Moon’s surface to identify resources, map terrain, and support future landing missions.

Additionally, the service will provide Space Domain Awareness (SDA). The AI software will fuse multiple data feeds to track maneuvering objects in cislunar space, the area between Earth and the Moon. This autonomous reconnaissance provides critical situational awareness for national security and safe space operations.

“Ocula is set to be the first commercial lunar imaging and mapping service available on the market,” said Jason Kim, CEO of Firefly Aerospace. “Now through our collaboration with NVIDIA, Ocula will be powered by the world’s leading edge AI processor. This capability allows us to layer on our SciTec AI software as the ‘brains’ that give customers real-time data driven insights from the Moon.”

AirPro News analysis

We view this development as a significant milestone in the commercialization of lunar infrastructure. Current government-owned lunar orbiters, such as NASA’s Lunar Reconnaissance Orbiter launched in 2009, are aging and nearing the end of their operational lives. Firefly’s Ocula service steps into a critical market void, offering updated, high-resolution lunar mapping to support the growing number of international and commercial lunar missions.

Furthermore, the emphasis on Space Domain Awareness highlights the growing strategic importance of cislunar space. As the Moon becomes more crowded with international missions, the ability to track maneuvering objects is a major priority for defense agencies like the U.S. Space Force. Firefly’s vertical integration, utilizing its own Elytra spacecraft, Blue Ghost lander, and SciTec software, demonstrates its maturation as an end-to-end space and defense contractor.

Frequently Asked Questions

What is the Ocula service?

Ocula is a commercial lunar imaging and mapping service developed by Firefly Aerospace. It utilizes NVIDIA edge AI technology to process high-resolution images directly in lunar orbit, sending actionable insights back to Earth.

When will the Ocula service launch?

The service will be deployed on Firefly’s Elytra orbital vehicle during Blue Ghost Mission 2, which is targeted for launch no earlier than late 2026.

What is Space Domain Awareness (SDA)?

SDA involves tracking and monitoring maneuvering objects in space. In this context, Ocula will monitor cislunar space (the area between Earth and the Moon) to provide situational awareness for safe space operations and national security.

Sources: Firefly Aerospace Official Press Release

This article is based on an official press release from Lockheed Martin.

On April 1, 2026, NASA’s Artemis II mission successfully launched, sending humans toward the Moon for the first time in over 50 years. As of this writing, the four-person crew is executing the latter half of their historic 10-day journey. At the heart of this mission is the Orion spacecraft, built by Lockheed Martin, which serves as the critical life-support vessel for the astronauts navigating the unforgiving environment of deep space.

According to an official press release from Lockheed Martin, the primary engineering focus of the Artemis II flight is the rigorous validation of Orion’s Environmental Control and Life Support System (ECLSS). This complex network of subsystems is actively keeping the crew alive, healthy, and comfortable as they travel on a hybrid free-return trajectory around the far side of the Moon.

Engineering Human Survival in Deep Space

The ECLSS is described by Lockheed Martin as the “first core function” of the Orion spacecraft. To make this mission a reality, manufacturers faced the monumental task of miniaturizing life support systems, which occupy massive amounts of space on the International Space Station (ISS), to fit within the strict size and mass limits of the Orion capsule without sacrificing efficacy.

Air, Water, and Thermal Control

As detailed in the company’s release, the Air Revitalization System utilizes regenerative chemical scrubbing technology called “amine swing beds” to maintain breathable oxygen, remove carbon dioxide, and control humidity. Crucially, in the event of a pressure vessel leak, this system can provide a pressurized, breathable atmosphere and thermal cooling for four suited astronauts for up to 144 hours.

The spacecraft’s Active Thermal Control System acts much like a car radiator, using coolant fluids and heat exchangers to vent excess heat into space. This ensures the cabin remains at a stable 70 to 75 degrees Fahrenheit, protecting the crew from the extreme temperature fluctuations of deep space and the intense heat of atmospheric reentry. Additionally, the Potable Water System supplies 74 gallons of highly filtered water across four pressurized tanks for drinking, hygiene, and medical needs.

Waste Management and Safety

Orion is equipped with a Universal Waste Management System modeled after the ISS space toilet. According to Lockheed Martin, it utilizes dual fan separators for airflow-assisted collection in microgravity, alongside advanced filtration for odor and particulate control. The spacecraft also features a dedicated hygiene bay for privacy and a microgravity-engineered Fire Detection and Suppression System that continuously monitors for combustion byproducts.

Rigorous Testing for a Historic Mission

Before the April 2026 launch, the ECLSS underwent exhaustive testing to ensure flawless operation in the vacuum of space. Lockheed Martin noted that hardware and “physics-only” tests were conducted at the Orion Life Support Integration Facility (OLIF) at NASA’s Johnson Space Center. Here, engineers simulated vacuum conditions, tested swing-bed seals, and verified pressure and humidity control loops under fault conditions.

Software validation took place at the Integrated Test Lab (ITL) near Denver, Colorado. Engineers ran full-mission simulations, injecting artificial faults, such as sensor noise or stuck valves, to test the system’s automated diagnostics and resolution capabilities.

“The Environmental Control and Life Support System is logically the first core function of Orion. Designing the spacecraft right is starting from the people onboard and working outward,” stated Sean O’Dell, Orion Spacecraft Architect at Lockheed Martin.

AirPro News analysis

We view the 144-hour emergency life support capability as one of the most critical engineering achievements of the Orion program. This robust safety net underscores the inherent, unforgiving dangers of deep space travel. By successfully stress-testing these systems in a true deep-space environment during Artemis II, NASA and Lockheed Martin are laying the essential groundwork for the lunar surface landings planned for Artemis III and, ultimately, future crewed missions to Mars.

Frequently Asked Questions

Who is on the Artemis II crew?

The historic four-person crew includes NASA astronauts Reid Wiseman (Commander), Victor Glover (Pilot), Christina Hammock Koch (Mission Specialist), and Canadian Space Agency astronaut Jeremy Hansen (Mission Specialist).

What is the primary goal of Artemis II?

Following the uncrewed Artemis I mission in 2022, Artemis II is a 10-day crewed flight test designed to validate Orion’s life support systems in a deep-space environment before future lunar surface landings.

How does Orion manage extreme temperatures?

Orion uses an Active Thermal Control System with coolant fluids and heat exchangers to absorb and vent excess heat, keeping the internal cabin at a comfortable 70 to 75 degrees Fahrenheit.

Sources

• Lockheed Martin

Introduction to the Next Generation of Satellite Operations

As the orbital environment becomes increasingly crowded, the management of satellite networks is undergoing a necessary technological evolution. On April 6, 2026, The Aerospace Corporation and Google Public Sector announced a joint initiative to modernize satellite operations through the integration of agentic artificial intelligence (AI). According to the official press release, the two organizations have co-developed a proof-of-concept tool designed to assist engineers and operators in managing the escalating complexity of Proliferated Low Earth Orbit (pLEO) constellations.

These pLEO networks, which consist of extensive arrays of small satellites operating at altitudes below 2,000 kilometers, are critical for enabling faster global communication and enhanced coverage. However, as these constellations scale from dozens to thousands of assets, the sheer volume and velocity of data generated can overwhelm traditional monitoring systems. The newly announced collaboration aims to address this bottleneck by leveraging Google Cloud’s Vertex AI platform to create an intelligent, unified interface for space system operators.

The Challenge of Scaling pLEO Constellations

Data Overload in Modern Space Operations

The rapid expansion of pLEO constellations has introduced unprecedented challenges for space system operators. According to the provided background data, managing these massive networks requires complex integration that stretches the limits of human monitoring. Current standard processes often force on-call engineers to manually rotate between disparate screens to track bus telemetry, payload status, and ground network availability.

When an anomaly occurs in orbit, operators typically lose critical minutes manually correlating this fragmented data to determine whether an alarm represents a genuine threat or a false positive. In an environment where split-second decisions are vital, this reactive approach leaves critical signals vulnerable to being lost in the noise of high-velocity data streams.

Agentic AI as a Force Multiplier

Predictive Behavioral Monitoring

To resolve the inefficiencies of manual data correlation, Aerospace collaborated with Google Public Sector’s Rapid Innovation Team to develop a tool that acts as a force multiplier. As detailed in the press release, the system utilizes agentic AI, an advanced form of artificial intelligence capable of reasoning and executing tasks autonomously, to automatically monitor the status of every satellite in a constellation.

By fusing disparate telemetry streams into a single interface, the tool shifts satellite management from static threshold alarms to predictive behavioral monitoring. The AI-based solution augments passive monitoring with active machine learning insights, allowing it to detect subtle behavioral anomalies before a component failure occurs. For instance, the system can identify nuances such as a momentum wheel oscillating only when a specific payload is active. Furthermore, the tool instantly correlates these anomalies with relevant external contexts, such as recent payload tasks or space weather events, presenting operators with an immediate root-cause analysis.

“This concept demonstrates how AI can be a critical operational partner capable of handling the high-velocity demands of modern space domain awareness, helping on-call engineers focus their expertise where it matters most,” said Kevin Bell, senior vice president of Aerospace’s Engineering and Technology Group, in the official announcement.

“This pathfinding effort demonstrates that by equipping engineers with the right data at the right time, we can help transform the operator experience from reactive firefighting to proactive problem-solving to accelerate operations,” added Cameron Groves, director of Rapid Innovation & Specialist Engineering at Google Public Sector.

Historical Context and Industry Trends

Building on Past Collaborations

This latest announcement builds upon a history of collaboration between the two organizations. In January 2025, Aerospace and Google Public Sector partnered on a groundbreaking initiative to revolutionize space weather forecasting. By combining Aerospace’s deep space science expertise with Google Cloud’s Vertex AI and high-performance computing, they developed a system capable of predicting geomagnetic storms days in advance, thereby safeguarding satellite communications and terrestrial power grids.

The Aerospace Corporation, a national nonprofit employing over 4,800 people, operates the only federally funded research and development center (FFRDC) dedicated to the space enterprise. Their ongoing partnership with Google Public Sector highlights a broader industry trend: the necessary fusion of commercial tech innovation with national security space programs to manage the growing complexities of the orbital domain.

AirPro News analysis

We view this development as a critical inflection point for Space Domain Awareness (SDA). As the space industry grapples with the “GenAI paradox” and the exponential growth of orbital objects, the computational expense and inherent vulnerabilities of manual satellite management are becoming unsustainable. The transition from reactive firefighting to proactive, AI-driven problem-solving is not merely an operational upgrade; it is a strategic necessity.

While this tool is currently in the proof-of-concept phase, its successful transition into operational use could set a new baseline for constellation management. By shifting the cognitive burden of data correlation from human engineers to agentic AI, organizations can ensure that their personnel are focused on high-level strategic decision-making. Ultimately, this pathfinding effort paves the way for the future development of fully autonomous, self-healing satellite networks, which will be essential for maintaining resilience in a congested and contested space environment.

Frequently Asked Questions (FAQ)

What is a pLEO constellation?

Proliferated Low Earth Orbit (pLEO) constellations are large-scale networks consisting of hundreds or thousands of small satellites orbiting at altitudes below 2,000 kilometers. They are designed to provide resilient, low-latency global connectivity for both commercial and defense applications.

What is agentic AI?

Agentic AI is an advanced form of artificial intelligence that utilizes large language models (LLMs) to reason, make decisions, and execute tasks autonomously. Unlike standard generative AI, agentic systems can use context and past experiences to respond to novel situations with minimal human intervention.

Sources

• The Aerospace Corporation Press Release