Artificial intelligence (AI) and machine learning (ML) are moving from terrestrial data centres into space-qualified electronic warfare (EW) systems, with Mercury Systems expanding its portfolio of radiation-tolerant processing platforms for missions spanning low Earth orbit (LEO), geostationary orbit (GEO) and deep space.

According to Mercury Systems, the transition towards onboard AI processing is reducing reliance on ground stations while increasing autonomous decision-making for defence and commercial spacecraft.

For manufacturers and programme developers, the move reflects continued demand for modular, space-qualified computing hardware that combines AI acceleration with radiation resilience and open architecture standards, reducing redesign requirements as mission capabilities evolve.

AI processing moves electronic warfare to the edge of space

According to the Mercury AI/ML blog, modern spacecraft increasingly require onboard processing to analyse sensor data, detect anomalies and respond without waiting for commands from Earth. This approach addresses the growing volume of Earth observation, intelligence, surveillance and reconnaissance (ISR), navigation and electronic warfare data generated in orbit.

As Mercury states, “AI-enabled EW systems are becoming essential components of modern space-based defence operations.” The company says onboard processing reduces latency, limits the need to transmit raw data and enables spacecraft to continue operating in contested or communication-denied environments.

The same source identifies several operational applications, including jamming detection, spectrum awareness, autonomous threat response, sensor fusion and signal intelligence processing.

Modular architectures support future hardware upgrades

Mercury attributes much of this development to Modular Open Systems Architecture (MOSA) standards. In its blog, the company highlights adoption of SpaceVPX, SOSA and OpenVPX standards to simplify integration of new processing technologies while avoiding complete hardware redesigns.

According to Mercury, these standards allow customers to “rapidly integrate new AI/ML capabilities” and replace computing elements as mission requirements change.

From a manufacturing perspective, open architectures can simplify long-term component selection by supporting hardware interoperability across multiple suppliers and programme lifecycles, particularly where missions remain operational for many years.

Radiation-tolerant AI hardware underpins autonomous missions

Mercury says modern AI workloads—including convolutional neural networks, transformer models and real-time inference—require substantially greater onboard processing capability than conventional spacecraft computers.

To address these requirements, the company identifies several technology building blocks:

  • Radiation-tolerant GPUs for AI and signal analysis.
  • Low-power neural processing units (NPUs) for autonomous inference.
  • Heterogeneous architectures combining CPUs, GPUs and FPGAs.

According to the company’s analysis, these processing platforms support image classification, signal analysis, autonomous navigation, electronic threat localisation and multi-sensor data fusion while operating under radiation exposure, thermal extremes and fault-tolerant requirements.

Mercury writes that these systems “must be engineered for radiation resilience, thermal management and fault tolerance.”

Space-qualified platforms extend from LEO to deep space

The company’s broader space technologies portfolio positions processing, storage, RF electronics and secure computing as core building blocks for space missions ranging from LEO constellations to deep-space exploration.

Within its AI blog, Mercury highlights its SCFE6933 SpaceVPX board, based on AMD Versal AI Core adaptive SoCs, as an example of onboard computing designed for machine learning inference, beamforming and software-defined radio applications. The company says the platform aligns with MOSA principles while supporting sensor fusion and mission processing across multiple orbital environments.

Mercury also describes practical applications across different mission profiles. LEO ISR satellites can perform onboard object recognition, GEO weather satellites can process large sensor datasets in orbit, while deep-space probes use machine learning for autonomous fault detection and recovery where communication delays prevent immediate human intervention.

The company’s strategy reflects a broader move towards higher-performance onboard computing capable of supporting increasingly autonomous spacecraft. For organisations planning future satellite programmes, the combination of open architectures, radiation-tolerant processing and AI acceleration is becoming a defining consideration in hardware selection, subsystem integration and long-term platform support.

Damian Semple, Franchise marketing manager, comments: “As AI workloads move onboard spacecraft, the focus extends beyond processing performance to securing long-term access to qualified components throughout a programme’s lifecycle. Early engagement with the supply chain and careful component selection can help reduce redesign risk and support production schedules as demand for space-grade computing hardware continues to grow.”

View the Mercury Systems product portfolio

For more electronics updates please visit: https://efemag.co.uk/category/news