Airfield automation is evolving quickly, and advancing these capabilities requires technology that can keep pace with the unpredictable conditions of open runways and nonstop operations. Autonomous ground vehicles need a computing platform that can see, think, and react in real time while enduring heat, vibration, and continuous environmental exposure. By integrating an industrial AI inference computer, the organization gained the reliable performance and rugged intelligence needed to move its airfield automation vision forward.
Challenges
- CPU performance limitations when running multi-model perception and autonomy pipelines
- Insufficient high-speed connectivity for sensors, cameras, and vehicle control networks
- Need for wide-temperature industrial memory and storage to ensure reliability in outdoor conditions
- Limited PCIe expansion capacity for GPU acceleration and future autonomy modules
- Requirement for a system that supports UL certification for deployment in regulated airfield environments
Solution
- Premio’s AI edge inference computer (RCO-6000-RPL)
- 12th/13th Gen Intel® TE processor support
- 32GB wide-temp DDR5 memory
- Supports 1TB wide-temp M.2 NVMe storage
- Eight USB 3.2 ports for strong high-speed sensor connectivity
- Hot-swappable 256GB wide-temp 2.5" SATA SSD
-
EDGEBoost I/O supporting:
- 2-Port RJ45 10GbE with Intel X710-AT2
- 4-Port PoE RJ45 1GbE
- EDGEBoost Nodes for GPU expansion
Benefits
- Long 10-year product lifecycle enabled by TE-series processors
- Faster engineering support with Premio’s accessible Los Angeles-based team
Company Overview
Focused on modernizing airfield operations, the organization builds autonomous ground vehicles and AI-based perception tools that give operators clearer visibility and faster decision-making capabilities. Their technology supports essential tasks—from spotting debris to monitoring pavement health and transporting equipment—ensuring safer and more efficient movement across the airfield. With AI-driven automation gaining momentum in both aviation and defense, the team stands ready to broaden its impact.
The Challenges
CPU performance limitations when running multi-model perception and autonomy pipelines
Autonomous systems depend on running multiple AI models simultaneously for detection, classification, navigation, and safety functions. Earlier compute platforms struggled to maintain real-time inference, impacting operational responsiveness. Higher multicore performance was needed to keep pace with evolving perception demands.
Insufficient high-speed connectivity for sensors, cameras, and vehicle control networks
A dense array of cameras, LiDAR units, and navigation modules requires substantial bandwidth to deliver timely perception data. Limited USB ports and the absence of PoE support restricted the integration of additional sensors, high-resolution visual systems, and powered IP cameras. Enhancing connectivity was essential to support the organization’s expanding autonomy stack.
Need for wide-temperature industrial memory and storage to ensure reliability in outdoor conditions
Airfield environments expose computing systems to heat, vibration, dust, and moisture. Commercial memory and storage risked performance degradation and data loss under these conditions. Industrial wide-temperature components were required for continuous, mission-critical operation.
Limited PCIe expansion capacity for GPU acceleration and future autonomy modules
Advanced perception increasingly relies on GPU acceleration to process high-frame-rate imagery and large neural networks. Prior hardware lacked adequate PCIe bandwidth to support GPUs or additional modules. A scalable architecture was needed to support future autonomy growth.
Requirement for a system that supports UL certification for deployment in regulated airfield environments
Strict compliance standards govern equipment used around aircraft and sensitive operational zones. Hardware without proper certification delays deployment and complicates integration. A UL-compliant system ensured smoother approvals and greater operational confidence.
The Solution
Premio’s AI edge inference computer (RCO-6000-RPL)
This rugged industrial edge computing platform supports real-time AI inference, multi-sensor fusion, and continuous autonomous operation. Its wide-temperature construction and durable design ensure reliability in harsh airfield environments. UL-listed certification and scalable expansion make it ideal for emerging autonomy workloads.
These processors provide the multicore performance needed to run complex perception and autonomy pipelines simultaneously. Their efficiency and extended lifecycle ensure long-term deployment stability. The platform can maintain real-time responsiveness even under heavy processing loads.
32GB wide-temp DDR5 memory
High-speed DDR5 memory supports rapid data movement between sensors and inference engines. Its industrial temperature tolerance ensures dependable operation in fluctuating outdoor conditions. This configuration helps avoid bottlenecks during advanced perception workloads.
1TB wide-temp M.2 NVMe storage
The wide-temperature NVMe drive offers high-speed access to inference models, logs, and mission-critical datasets. Its rugged design withstands vibration and environmental stress common in airfield deployments. Native M.2 NVMe support ensures fast, reliable primary storage without expansion requirements.
Eight USB 3.2 ports for high-speed sensor connectivity
The system includes eight high-speed USB 3.2 ports without requiring additional expansion modules. This allows direct connection of high-resolution cameras, navigation peripherals, and diagnostic devices. Strong native connectivity ensures low-latency data flow for real-time perception.
Hot-swappable 256GB wide-temp 2.5" SATA SSD
The removable drive allows quick updates, replacements, and mission-data offloading without downtime. Industrial durability ensures resilience against vibration and temperature extremes. This provides a dependable secondary storage layer.
EDGEBoost I/O modules for enhanced networking
Modular expansion supports two dedicated networking options essential for high-bandwidth and PoE-powered devices: a 2-port RJ45 10GbE module with Intel X710-AT2 for heavy data pipelines, and a 4-port PoE RJ45 1GbE module for powering IP cameras and edge sensors. These modules provide flexible, scalable connectivity for increasingly complex autonomy stacks. Their integration ensures the system can support dense, diverse sensor ecosystems.
EDGEBoost Nodes for GPU expansion
GPU acceleration enhances real-time perception, mapping, and neural network inference, enabling richer environmental understanding and faster decision-making. Expansion capability ensures the platform can adapt as autonomy models grow more complex. This future-ready architecture supports next-generation AI workloads.
The Benefits
Long 10-year product lifecycle with TE-series processors
TE processors ensure extended availability, reducing redesign cycles and supporting stable long-term deployment for autonomous platforms.
Faster engineering support with Premio’s Los Angeles-based team
Proximity to engineering resources enables rapid assistance, integration support, and streamlined development collaboration.
Conclusion
The industrial AI inference computer gave the team the dependable performance and resilience they needed to push autonomous airfield operations forward. With powerful processing, rugged storage, expansive connectivity, and room for GPU-driven growth, the system now handles real-time perception with confidence. This foundation sets the stage for continued advancements in airfield intelligence and the future of automated operations.