Engineering safety into modern systems

by Jessica Brownlow | Apr 9, 2025 | Avionics, Innovation

Article by Gabi Davenport, Intel Corporation 

In today’s increasingly automated world, the systems we build aren’t just expected to work—they’re expected to work safely, often under extreme conditions and with little room for error. From industrial robotics on the factory floor to autonomous aircraft in our skies, safety isn’t an afterthought—it’s a design requirement.

Nowhere is this more evident than in aerospace, where flight control systems, navigation algorithms, and onboard computing platforms must operate with unwavering reliability. A transient hardware fault, if undetected, can compromise mission integrity or human safety. The challenge is similar across other sectors like autonomous vehicles, smart energy grids, and industrial automation.

As the hardware powering these systems becomes more complex, so do the challenges in ensuring their integrity.

Modern computing systems are built around highly integrated system-on-chip (SoC) architectures that combine compute cores, graphics units, memory, communication fabrics, and power management—all on a single chip. While this integration brings performance and power efficiency, it also increases the surface area for potential faults.

In safety-critical applications like aerospace, the stakes are particularly high. Systems must:

• Continuously monitor for faults, including transient radiation-induced errors (common at altitude),
• Run health diagnostics without interrupting operation,
• Recover gracefully from detected issues,
• Comply with strict standards like DO-254, IEC 61508, or ISO 13849.

Building this resilience requires a shift from reactive fault handling to proactive fault prevention and detection, starting from the silicon layer up.

A recent white paper from Intel (below) highlights one example of this design philosophy in action. The Silicon Integrity Technology outlines integrated features embedded into select Intel® processors to support functional safety.

While Intel’s technology targets a wide range of markets, including industrial automation and autonomous systems, its features map directly to the needs of aerospace computing:

• Dual-Core Lockstep (DCLS): Two cores run identical operations to detect faults, crucial for avionics and flight-critical functions.
• In-Field Diagnostics: Performs self-tests at startup or during operation, which is vital for aerospace systems requiring uninterrupted self-checks.
• Error Detection and Correction (ECC): Protects against radiation-induced errors in high-altitude environments, ensuring data accuracy.
• Integrated Monitoring and Logging: Records and reports faults through standard interfaces, essential for real-time fault management in aerospace systems.

Notably, these capabilities are managed through system firmware and bootloaders, allowing flexibility in how and when they’re used based on mission profiles.

The need for high reliability under constrained conditions is particularly acute in aerospace. Systems often operate in radiation-prone environments, with limited access for maintenance, and must meet strict certification requirements. Integrating diagnostic and error-handling features directly into the processor helps address many pain points.

Engineers can now leverage integrity-aware silicon to build smarter, more efficient safety architectures rather than relying solely on redundant systems (which add cost, weight, and complexity). This shift opens new doors for applications like autonomous drones, satellite systems, and next-generation avionics platforms.

Whether you’re designing for aerospace, industrial robotics, or autonomous vehicles, the goal is to reduce undetected failure rates, enhance fault coverage, and simplify the path to certification.

Conclusion

As embedded systems take on more autonomous roles in our lives and industries, the demand for safety-integrated hardware will only grow. Tools like Intel® Silicon Integrity Technology are helping shift the paradigm—embedding fault detection and system integrity features into the fabric of the processor itself.

These hardware-level innovations will be key to building tomorrow’s safe, resilient systems—whether on the ground, in the factory, or 35,000 feet above it.

For more like this see:

• Is aerospace ready for quantum computing?
• Reskilling for the AI era: How gen AI is changing the industry landscape
• Summary: The UK Civil Aviation Authority’s new AI strategy